Loading video...

Video Failed to Load

Go Home

Check out our #RSS2024 paper "#MPCC++: Model Predictive Contouring Control for Time-Optimal Flight with Safety Constraints." Model Predictive Contouring Control (MPCC) has shown promising results for agile robotics applications, including car and drone racing. Existing approaches struggle to introduce safety considerations, often resulting in crashes. What does it take...

davsca1's profile picture

Davide Scaramuzza

18,362 subscribers

17,903 views • 2 years ago •via X (Twitter)

Science & TechnologyEducationHealth & Wellness#RSS2024#MPCC
LIVE

Anya Rossi• Live Now

Private livecam show

3 Comments

Julian Fried's profile picture
Julian Fried2 years ago

Very cool

Charles Zhang's profile picture
Charles Zhang2 years ago

cool

Senish Khadka's profile picture
Senish Khadka2 years ago

Damn, but how did you localize the drones position? Imu?

Related Videos

Check out our #ICRA2024 paper "Actor-Critic Model Predictive Control." Model-free #reinforcementlearning (RL) is known for its strong task performance and flexibility in optimizing general reward formulations. On the other hand, #ModelPredictiveControl (MPC) benefits from robustness and online replanning capabilities. We combine both approaches by introducing a new framework called Actor-Critic Model Predictive Control. The key idea is to embed a differentiable MPC within an Actor-Critic RL framework. The proposed approach leverages the short-term predictive optimization capabilities of MPC with the exploratory and end-to-end training properties of RL. The resulting policy effectively manages both short-term decisions through the MPC-based actor and long-term prediction via the critic network, unifying the benefits of both model-based control and end-to-end learning. We validate our method in simulation and the real world with a quadcopter across various high-level tasks. We show that the proposed architecture can achieve real-time control performance, learn complex behaviors via trial and error, and retain the predictive properties of the MPC to better handle out-of-distribution behavior. Paper: Full Video with more details: Kudos to Ángel Romero, Yunlong Song IEEE ICRA University of Zurich UZH Science UZH Space Hub Aerial Core AUTOASSESS European Research Council (ERC)
1:26

Check out our #ICRA2024 paper "Actor-Critic Model Predictive Control." Model-free #reinforcementlearning (RL) is known for its strong task performance and flexibility in optimizing general reward formulations. On the other hand, #ModelPredictiveControl (MPC) benefits from robustness and online replanning capabilities. We combine both approaches by introducing a new framework called Actor-Critic Model Predictive Control. The key idea is to embed a differentiable MPC within an Actor-Critic RL framework. The proposed approach leverages the short-term predictive optimization capabilities of MPC with the exploratory and end-to-end training properties of RL. The resulting policy effectively manages both short-term decisions through the MPC-based actor and long-term prediction via the critic network, unifying the benefits of both model-based control and end-to-end learning. We validate our method in simulation and the real world with a quadcopter across various high-level tasks. We show that the proposed architecture can achieve real-time control performance, learn complex behaviors via trial and error, and retain the predictive properties of the MPC to better handle out-of-distribution behavior. Paper: Full Video with more details: Kudos to Ángel Romero, Yunlong Song IEEE ICRA University of Zurich UZH Science UZH Space Hub Aerial Core AUTOASSESS European Research Council (ERC)

Davide Scaramuzza

34,889 views • 2 years ago

Check out our latest work, "Actor-Critic Model Predictive Control: Differentiable Optimization meets Reinforcement Learning for Agile Flight," published in the IEEE Transactions on Robotics, where we reconcile #OptimalControl and #ReinforcementLearning, achieving the same super-human performance, but with superior generalizability, as our previous model-free deep RL! Code released! PDF: Code: Full Video: Model-free #ReinforcementLearning (RL) is known for its strong task performance and flexibility in optimizing general reward formulations. On the other hand, #ModelPredictiveControl (MPC) provides robustness, constraint handling, and powerful online replanning capabilities. In this work, we extend our previous AC-MPC paper (Romero, ICRA'24) by taking a deeper look at how both approaches can be unified. We introduce and extend Actor-Critic Model Predictive Control (AC-MPC), a framework that embeds a differentiable MPC inside an Actor-Critic RL architecture. This integration allows the MPC-based actor to perform short-term predictive optimization, while the critic facilitates long-horizon learning and exploration. We conduct a comprehensive study that highlights AC-MPC’s key advantages: - Better out-of-distribution generalization, both against unknown disturbances and changes in the quadrotor dynamics - Improved sample efficiency - A novel empirical analysis uncovering a relationship between the critic’s value function and the MPC cost function, providing deeper insight into their interplay. We validate our method in simulation and the real world on a quadcopter flying at superhuman speeds of up to 21 m/s, matching state-of-the-art model-free RL performance, and retaining the predictive structure of MPC for more reliable out-of-distribution behavior. Reference: Actor-Critic Model Predictive Control: Differentiable Optimization meets Reinforcement Learning for Agile Flight IEEE Transactions on Robotics (T-RO), 2025 PDF: Full Video: Code: Kudos to Ángel Romero, Elie Aljalbout, Yunlong Song! University of Zurich UZH Science UZH Space Hub AUTOASSESS European Research Council (ERC) UZHai
1:46

Check out our latest work, "Actor-Critic Model Predictive Control: Differentiable Optimization meets Reinforcement Learning for Agile Flight," published in the IEEE Transactions on Robotics, where we reconcile #OptimalControl and #ReinforcementLearning, achieving the same super-human performance, but with superior generalizability, as our previous model-free deep RL! Code released! PDF: Code: Full Video: Model-free #ReinforcementLearning (RL) is known for its strong task performance and flexibility in optimizing general reward formulations. On the other hand, #ModelPredictiveControl (MPC) provides robustness, constraint handling, and powerful online replanning capabilities. In this work, we extend our previous AC-MPC paper (Romero, ICRA'24) by taking a deeper look at how both approaches can be unified. We introduce and extend Actor-Critic Model Predictive Control (AC-MPC), a framework that embeds a differentiable MPC inside an Actor-Critic RL architecture. This integration allows the MPC-based actor to perform short-term predictive optimization, while the critic facilitates long-horizon learning and exploration. We conduct a comprehensive study that highlights AC-MPC’s key advantages: - Better out-of-distribution generalization, both against unknown disturbances and changes in the quadrotor dynamics - Improved sample efficiency - A novel empirical analysis uncovering a relationship between the critic’s value function and the MPC cost function, providing deeper insight into their interplay. We validate our method in simulation and the real world on a quadcopter flying at superhuman speeds of up to 21 m/s, matching state-of-the-art model-free RL performance, and retaining the predictive structure of MPC for more reliable out-of-distribution behavior. Reference: Actor-Critic Model Predictive Control: Differentiable Optimization meets Reinforcement Learning for Agile Flight IEEE Transactions on Robotics (T-RO), 2025 PDF: Full Video: Code: Kudos to Ángel Romero, Elie Aljalbout, Yunlong Song! University of Zurich UZH Science UZH Space Hub AUTOASSESS European Research Council (ERC) UZHai

Davide Scaramuzza

27,090 views • 6 months ago

We are excited to share that our paper “Low-Latency Event-Based Velocimetry for Quadrotor Control in a Narrow Pipe” published in the IEEE Transactions on Robotics! One year in the making! PDF: Video: We focus on autonomous quadrotor flight in narrow pipes, where self-induced, unsteady airflow can significantly affect stability and control. Our approach closes the loop using real-time flow field measurements: an #EventCamera-based smoke velocimetry method provides low-latency airflow estimates, which are used by a learning-based disturbance estimator and integrated into a reinforcement-learning controller. The results are improved hovering and lateral translation performance, helping make flight in confined spaces more stable and safer! Kudos to Leonard Bauersfeld Leonard Bauersfeld! Reference: Low-Latency Event-Based Velocimetry for Quadrotor Control in a Narrow Pipe, IEEE Transactions on Robotics (T-RO), 2026 European Research Council (ERC) UZH IfI UZH Space Hub University of Zurich Swiss Robotics UZH Science UZHai AUTOASSESS IEEE Transactions on Robotics (T-RO) #DVS Prophesee
2:11

We are excited to share that our paper “Low-Latency Event-Based Velocimetry for Quadrotor Control in a Narrow Pipe” published in the IEEE Transactions on Robotics! One year in the making! PDF: Video: We focus on autonomous quadrotor flight in narrow pipes, where self-induced, unsteady airflow can significantly affect stability and control. Our approach closes the loop using real-time flow field measurements: an #EventCamera-based smoke velocimetry method provides low-latency airflow estimates, which are used by a learning-based disturbance estimator and integrated into a reinforcement-learning controller. The results are improved hovering and lateral translation performance, helping make flight in confined spaces more stable and safer! Kudos to Leonard Bauersfeld Leonard Bauersfeld! Reference: Low-Latency Event-Based Velocimetry for Quadrotor Control in a Narrow Pipe, IEEE Transactions on Robotics (T-RO), 2026 European Research Council (ERC) UZH IfI UZH Space Hub University of Zurich Swiss Robotics UZH Science UZHai AUTOASSESS IEEE Transactions on Robotics (T-RO) #DVS Prophesee

Davide Scaramuzza

50,192 views • 5 months ago

We are thrilled to share our breakthrough research on "Agile Flight from Pixels without State Estimation," to be presented and live-demonstrated at #RSS2024 next week! You heard well: no state estimation means no explicit visual localization, no SLAM, no VIO, and no IMU! Paper: Video (Narrated): Last year, we demonstrated that #ReinforcementLearning (RL) policies could outperform world-champion drone-racing pilots using the same quadrotor hardware; however, unlike human pilots, these policies continuously estimated an explicit state from known gate positions, the camera feed, and inertial measurements (IMU). In this new work, we tackle the challenge of learning vision-based drone racing using an end-to-end reinforcement learning approach that eliminates the need for IMU data or explicit state estimation. Like professional pilots, we go directly from images to control commands. The training is facilitated by an asymmetric actor-critic with access to privileged information. To overcome the computational complexity during image-based RL training, we use an appropriate sensor representation, which can be efficiently simulated during training without rendering images. We achieve agile flight at speeds up to 40 km/h with accelerations up to 2 g's. Although our demonstration focuses on drone racing, we believe that our method has an impact beyond drone racing and can serve as a foundation for future research into real-world applications in structured environments. Besides the paper presentation, we will also give a live demo next Tuesday and Wednesday between and hrs at TU Delft: Reference: Ismail Geles*, Leonard Bauersfeld*, Angel Romero, Jiaxu Xing, Davide Scaramuzza "Demonstrating Agile Flight from Pixels without State Estimation" Robotics: Science and Systems (RSS), 2024. Kudos to Ismail Geles Leonard Bauersfeld Ángel Romero Jiaxu Xing! University of Zurich UZH Science UZH Space Hub Aerial Core AUTOASSESS European Research Council (ERC)
1:24

We are thrilled to share our breakthrough research on "Agile Flight from Pixels without State Estimation," to be presented and live-demonstrated at #RSS2024 next week! You heard well: no state estimation means no explicit visual localization, no SLAM, no VIO, and no IMU! Paper: Video (Narrated): Last year, we demonstrated that #ReinforcementLearning (RL) policies could outperform world-champion drone-racing pilots using the same quadrotor hardware; however, unlike human pilots, these policies continuously estimated an explicit state from known gate positions, the camera feed, and inertial measurements (IMU). In this new work, we tackle the challenge of learning vision-based drone racing using an end-to-end reinforcement learning approach that eliminates the need for IMU data or explicit state estimation. Like professional pilots, we go directly from images to control commands. The training is facilitated by an asymmetric actor-critic with access to privileged information. To overcome the computational complexity during image-based RL training, we use an appropriate sensor representation, which can be efficiently simulated during training without rendering images. We achieve agile flight at speeds up to 40 km/h with accelerations up to 2 g's. Although our demonstration focuses on drone racing, we believe that our method has an impact beyond drone racing and can serve as a foundation for future research into real-world applications in structured environments. Besides the paper presentation, we will also give a live demo next Tuesday and Wednesday between and hrs at TU Delft: Reference: Ismail Geles*, Leonard Bauersfeld*, Angel Romero, Jiaxu Xing, Davide Scaramuzza "Demonstrating Agile Flight from Pixels without State Estimation" Robotics: Science and Systems (RSS), 2024. Kudos to Ismail Geles Leonard Bauersfeld Ángel Romero Jiaxu Xing! University of Zurich UZH Science UZH Space Hub Aerial Core AUTOASSESS European Research Council (ERC)

Davide Scaramuzza

27,917 views • 2 years ago

We are excited to share our #ICRA2026 paper "Dream to Fly: Model-Based Reinforcement Learning for Vision-Based Drone Flight"! Paper: Video: Can we use Model-Based #ReinforcementLearning (MBRL) to fly a drone from pixels to commands? In this work, we train quadrotor navigation policies from scratch using #WorldModels, mapping raw onboard camera pixels directly to control commands, much like a human pilot! While model-free methods like PPO are sample-inefficient and struggle in this setting, we leverage #MBRL to train visuomotor policies capable of agile flight through a racetrack using only raw pixel observations, no explicit state estimation needed. A key finding: because our policies are trained end-to-end directly from pixels, we no longer need the perception-aware reward term used in previous methods. Instead, this behavior emerges naturally! The policies learn to guide the camera toward feature-rich areas of the observation space on their own. Kudos to Ángel Romero Ashwin Shenai Ismail Geles Elie Aljalbout Reference: "Dream to Fly: Model-Based Reinforcement Learning for Vision-Based Drone Flight" Angel Romero*, Ashwin Shenai*, Ismail Geles, Elie Aljalbout, Davide Scaramuzza IEEE International Conference on Robotics and Automation (ICRA), Vienna, 2026. European Research Council (ERC) AUTOASSESS UZH IfI University of Zurich UZH Science Prophesee SynSense UZH Space Hub Swiss Robotics NCCR Robotics
1:21

We are excited to share our #ICRA2026 paper "Dream to Fly: Model-Based Reinforcement Learning for Vision-Based Drone Flight"! Paper: Video: Can we use Model-Based #ReinforcementLearning (MBRL) to fly a drone from pixels to commands? In this work, we train quadrotor navigation policies from scratch using #WorldModels, mapping raw onboard camera pixels directly to control commands, much like a human pilot! While model-free methods like PPO are sample-inefficient and struggle in this setting, we leverage #MBRL to train visuomotor policies capable of agile flight through a racetrack using only raw pixel observations, no explicit state estimation needed. A key finding: because our policies are trained end-to-end directly from pixels, we no longer need the perception-aware reward term used in previous methods. Instead, this behavior emerges naturally! The policies learn to guide the camera toward feature-rich areas of the observation space on their own. Kudos to Ángel Romero Ashwin Shenai Ismail Geles Elie Aljalbout Reference: "Dream to Fly: Model-Based Reinforcement Learning for Vision-Based Drone Flight" Angel Romero*, Ashwin Shenai*, Ismail Geles, Elie Aljalbout, Davide Scaramuzza IEEE International Conference on Robotics and Automation (ICRA), Vienna, 2026. European Research Council (ERC) AUTOASSESS UZH IfI University of Zurich UZH Science Prophesee SynSense UZH Space Hub Swiss Robotics NCCR Robotics

Davide Scaramuzza

15,986 views • 3 months ago

Check out our #ICRA2024 paper "Contrastive Initial State Buffer for Reinforcement Learning," which tackles the sample inefficiency in #ReinforcementLearning head-on. Code released! We introduce an approach agnostic to the underlying RL algorithm: the Contrastive Initial State Buffer. This tool strategically selects states from past experiences and uses them to initialize the agent in the environment to guide it toward more informative states. Our experiments on drone racing and legged locomotion show that our method achieves higher task performance while also speeding up training convergence. Reference: Nico Messikommer, Yunlong Song, Davide Scaramuzza Contrastive Initial State Buffer for Reinforcement Learning IEEE International Conference on Robotics and Automation (ICRA), 2024. PDF: Code: Video: Kudos to Messikommer Yunlong Song Aerial Core European Research Council (ERC) University of Zurich UZH Space Hub IEEE ICRA UZH Science
1:12

Check out our #ICRA2024 paper "Contrastive Initial State Buffer for Reinforcement Learning," which tackles the sample inefficiency in #ReinforcementLearning head-on. Code released! We introduce an approach agnostic to the underlying RL algorithm: the Contrastive Initial State Buffer. This tool strategically selects states from past experiences and uses them to initialize the agent in the environment to guide it toward more informative states. Our experiments on drone racing and legged locomotion show that our method achieves higher task performance while also speeding up training convergence. Reference: Nico Messikommer, Yunlong Song, Davide Scaramuzza Contrastive Initial State Buffer for Reinforcement Learning IEEE International Conference on Robotics and Automation (ICRA), 2024. PDF: Code: Video: Kudos to Messikommer Yunlong Song Aerial Core European Research Council (ERC) University of Zurich UZH Space Hub IEEE ICRA UZH Science

Davide Scaramuzza

13,846 views • 2 years ago

We are excited to share our latest work, "Learning on the Fly: Rapid Policy Adaptation via Differentiable Simulation", where a policy learns to adapt in the real world to unknown disturbances within 5 seconds, both with and without explicit state estimation, directly from visual features. Code released! PDF: Project Page: Starting from a simple analytical dynamics model, the system continuously learns residual dynamics from real-world data and embeds the refined model into a differentiable simulator. This enables fast, gradient-based policy updates that are far more sample-efficient than classical #ReinforcementLearning. We demonstrate rapid adaptation in <5 seconds in agile quadrotor control under challenging conditions, including added payloads, wind disturbances, and large sim-to-real gaps. In real-world experiments, our method reduces hovering error by up to 81% compared to L1-MPC and 55% compared to PPO-based adaptive methods. It also operates directly from visual features without explicit state estimation. Reference: “Learning on the Fly: Rapid Policy Adaptation via Differentiable Simulation” IEEE Robotics and Automation Letters, 2026 PDF: Video: Code: Website: Kudos to Michael Pan, Jiaxu Xing, Rudolf Reiter, Yifan Zhai, Elie Aljalbout! UZH Space Hub UZH IfI European Research Council (ERC) AUTOASSESS UZH Science University of Zurich
1:06

We are excited to share our latest work, "Learning on the Fly: Rapid Policy Adaptation via Differentiable Simulation", where a policy learns to adapt in the real world to unknown disturbances within 5 seconds, both with and without explicit state estimation, directly from visual features. Code released! PDF: Project Page: Starting from a simple analytical dynamics model, the system continuously learns residual dynamics from real-world data and embeds the refined model into a differentiable simulator. This enables fast, gradient-based policy updates that are far more sample-efficient than classical #ReinforcementLearning. We demonstrate rapid adaptation in <5 seconds in agile quadrotor control under challenging conditions, including added payloads, wind disturbances, and large sim-to-real gaps. In real-world experiments, our method reduces hovering error by up to 81% compared to L1-MPC and 55% compared to PPO-based adaptive methods. It also operates directly from visual features without explicit state estimation. Reference: “Learning on the Fly: Rapid Policy Adaptation via Differentiable Simulation” IEEE Robotics and Automation Letters, 2026 PDF: Video: Code: Website: Kudos to Michael Pan, Jiaxu Xing, Rudolf Reiter, Yifan Zhai, Elie Aljalbout! UZH Space Hub UZH IfI European Research Council (ERC) AUTOASSESS UZH Science University of Zurich

Davide Scaramuzza

19,184 views • 6 months ago

We are excited to share our #CORL2024 paper (oral) on "Learning Quadruped Locomotion Using Differentiable Simulation" done in collaboration with Sangbae Kim Massachusetts Institute of Technology (MIT). We present a new way to learn to walk in minutes without parallelization, outperforming PPO in sample efficiency! PDF: Video: We present a new framework for learning quadruped locomotion. By leveraging differentiable simulation for policy optimization, our approach achieves fast convergence and stable training, significantly outperforming model-free #ReinforcementLearning methods like PPO in sample efficiency. The key enabler is to combine a high-fidelity, non-differentiable simulator for forward dynamics with a simplified surrogate model for gradient backpropagation. Our framework enables learning quadruped walking in simulation in minutes without parallelization. When augmented with GPU parallelization, our approach allows the quadruped robot to master diverse locomotion skills on challenging terrains in minutes. This work highlights one of the first successful real-world applications of differentiable simulation for quadruped robots, offering a compelling alternative to traditional RL methods. Kudos to Yunlong Song! UZH Science University of Zurich UZH Space Hub UZH IfI European Research Council (ERC) Massachusetts Institute of Technology (MIT)MechE
1:13

We are excited to share our #CORL2024 paper (oral) on "Learning Quadruped Locomotion Using Differentiable Simulation" done in collaboration with Sangbae Kim Massachusetts Institute of Technology (MIT). We present a new way to learn to walk in minutes without parallelization, outperforming PPO in sample efficiency! PDF: Video: We present a new framework for learning quadruped locomotion. By leveraging differentiable simulation for policy optimization, our approach achieves fast convergence and stable training, significantly outperforming model-free #ReinforcementLearning methods like PPO in sample efficiency. The key enabler is to combine a high-fidelity, non-differentiable simulator for forward dynamics with a simplified surrogate model for gradient backpropagation. Our framework enables learning quadruped walking in simulation in minutes without parallelization. When augmented with GPU parallelization, our approach allows the quadruped robot to master diverse locomotion skills on challenging terrains in minutes. This work highlights one of the first successful real-world applications of differentiable simulation for quadruped robots, offering a compelling alternative to traditional RL methods. Kudos to Yunlong Song! UZH Science University of Zurich UZH Space Hub UZH IfI European Research Council (ERC) Massachusetts Institute of Technology (MIT)MechE

Davide Scaramuzza

15,533 views • 1 year ago

We are excited to share our latest work, "Superhuman Safe and Agile Racing through Multi-Agent Reinforcement Learning," done in collaboration with Google DeepMind . Autonomous drones have reached superhuman speed in isolation, but what happens when multiple agents share the same airspace? Paper: Website: Video: Using league-based self-play, we train #ReinforcementLearning agents that race against a diverse, evolving population of opponents. Through this competitive training, sophisticated behaviors emerge without explicit programming: strategic overtaking, proactive collision avoidance, and even awareness of aerodynamic downwash from nearby drones. In real-world multi-player races at speeds exceeding 80kph (50 mph) and accelerations up to 7g, our agents outperform a five-time Swiss national drone racing champion while reducing collision rates by 50% compared to single-agent baselines. Crucially, training against diverse artificial opponents enables zero-shot generalization to human pilots, achieving over 90% race completion in mixed human-AI races with up to four competitors. A key insight: human pilots adopt riskier strategies when trailing, leading to more crashes under competitive pressure. Our learned policies, by contrast, maintain consistent safety margins regardless of race standing, a property essential for deploying autonomous systems alongside humans. Also, the multi-agent self-play policies are more robust than those trained independently, suggesting that training in competitive environments is not only key to winning races but also to learning safer, more reliable autonomy for real-world multi-robot systems. Kudos to Ismail Geles, Leonard Bauersfeld, Markus Wulfmeier! Ismail Geles Leonard Bauersfeld Markus Wulfmeier European Research Council (ERC) UZH IfI University of Zurich UZH Science UZH Space Hub Swiss Robotics NCCR Robotics
1:47

We are excited to share our latest work, "Superhuman Safe and Agile Racing through Multi-Agent Reinforcement Learning," done in collaboration with Google DeepMind . Autonomous drones have reached superhuman speed in isolation, but what happens when multiple agents share the same airspace? Paper: Website: Video: Using league-based self-play, we train #ReinforcementLearning agents that race against a diverse, evolving population of opponents. Through this competitive training, sophisticated behaviors emerge without explicit programming: strategic overtaking, proactive collision avoidance, and even awareness of aerodynamic downwash from nearby drones. In real-world multi-player races at speeds exceeding 80kph (50 mph) and accelerations up to 7g, our agents outperform a five-time Swiss national drone racing champion while reducing collision rates by 50% compared to single-agent baselines. Crucially, training against diverse artificial opponents enables zero-shot generalization to human pilots, achieving over 90% race completion in mixed human-AI races with up to four competitors. A key insight: human pilots adopt riskier strategies when trailing, leading to more crashes under competitive pressure. Our learned policies, by contrast, maintain consistent safety margins regardless of race standing, a property essential for deploying autonomous systems alongside humans. Also, the multi-agent self-play policies are more robust than those trained independently, suggesting that training in competitive environments is not only key to winning races but also to learning safer, more reliable autonomy for real-world multi-robot systems. Kudos to Ismail Geles, Leonard Bauersfeld, Markus Wulfmeier! Ismail Geles Leonard Bauersfeld Markus Wulfmeier European Research Council (ERC) UZH IfI University of Zurich UZH Science UZH Space Hub Swiss Robotics NCCR Robotics

Davide Scaramuzza

14,612 views • 2 months ago

Can an inexpensive, off-the-shelf IMU be the only sensor to estimate the full state (position, velocity, orientation) of a quadrotor flying through a track at high speed and even be on-pair with vision-based localization? The answer is yes, within certain limitations! In this #RAL2023 paper, we propose a learning-based odometry algorithm that couples a model-based filter driven by the inertial measurements with a learning-based module with access to the control commands. Our system outperforms by a large margin the state-of-the-art visual-inertial odometry (#VIO) algorithms and the state-of-the-art learned-inertial odometry algorithm, #TLIO, for the task of drone racing. Additionally, we show that our system is as accurate as a VIO algorithm that uses a camera to localize to a known map of the racing track. The main limitation of our approach is that it cannot generalize to trajectories that have not been seen at training time. However, in drone racing competitions, the track is known beforehand. Human pilots spend hours or even days of practice on the race track before the competition. Similarly, our system can be trained with the data collected during practice time and deployed during the competition. Future work will investigate how to generalize to trajectories not seen at training time. The code is released! Paper: Video: Code: Kudos to Giovanni Cioffi Leonard Bauersfeld Elia Kaufmann European Research Council (ERC) University of Zurich UZH Science UZH Space Hub NCCR Robotics Aerial Core #RAL2023 #IROS2023 #SLAM
0:31

Can an inexpensive, off-the-shelf IMU be the only sensor to estimate the full state (position, velocity, orientation) of a quadrotor flying through a track at high speed and even be on-pair with vision-based localization? The answer is yes, within certain limitations! In this #RAL2023 paper, we propose a learning-based odometry algorithm that couples a model-based filter driven by the inertial measurements with a learning-based module with access to the control commands. Our system outperforms by a large margin the state-of-the-art visual-inertial odometry (#VIO) algorithms and the state-of-the-art learned-inertial odometry algorithm, #TLIO, for the task of drone racing. Additionally, we show that our system is as accurate as a VIO algorithm that uses a camera to localize to a known map of the racing track. The main limitation of our approach is that it cannot generalize to trajectories that have not been seen at training time. However, in drone racing competitions, the track is known beforehand. Human pilots spend hours or even days of practice on the race track before the competition. Similarly, our system can be trained with the data collected during practice time and deployed during the competition. Future work will investigate how to generalize to trajectories not seen at training time. The code is released! Paper: Video: Code: Kudos to Giovanni Cioffi Leonard Bauersfeld Elia Kaufmann European Research Council (ERC) University of Zurich UZH Science UZH Space Hub NCCR Robotics Aerial Core #RAL2023 #IROS2023 #SLAM

Davide Scaramuzza

37,061 views • 2 years ago

We are excited to share our work “Event-Aided Sharp Radiance Field Reconstruction for Fast-Flying Drones” published in IEEE Transactions on Robotics IEEE Transactions on Robotics (T-RO), which tackles sharp radiance field reconstruction under agile drone motion, where RGB frames are heavily motion-blurred and pose priors become unreliable! 4 years in the making! Code & dataset released! PDF: Code & Dataset: Full Narrated Video: High-speed flight is essential for time- and battery-constrained missions (e.g., inspection, exploration, search & rescue). However, fast motion corrupts visual data with severe motion blur and introduces drift/noise in visual-inertial odometry, making NeRF-based 3D reconstruction particularly brittle. We propose a unified framework that leverages asynchronous #EventCamera streams together with motion-blurred frames to reconstruct high-fidelity radiance fields from agile drone flights. Our key idea is to embed event-image fusion directly into radiance field optimization while jointly refining a shared, continuous-time camera trajectory initialized from event-based VIO. This enables us to recover sharp radiance fields and accurate trajectories without ground-truth supervision during training. We validate our method on synthetic data and on real sequences captured by a drone flying up to 2 m/s. Despite severe blur and noisy pose priors, our method preserves fine scene details and achieves a performance gain of over 50% on real-world data compared to state-of-the-art methods. Kudos to Rong Zou and Marco Cannici! Marco Cannici Reference: Rong Zou*, Marco Cannici*, Davide Scaramuzza Event-Aided Sharp Radiance Field Reconstruction for Fast-Flying Drones IEEE Transactions on Robotics (T-RO), 2026 NCCR Robotics European Research Council (ERC) AUTOASSESS UZH IfI University of Zurich UZH Science Prophesee SynSense UZH Space Hub
1:02

We are excited to share our work “Event-Aided Sharp Radiance Field Reconstruction for Fast-Flying Drones” published in IEEE Transactions on Robotics IEEE Transactions on Robotics (T-RO), which tackles sharp radiance field reconstruction under agile drone motion, where RGB frames are heavily motion-blurred and pose priors become unreliable! 4 years in the making! Code & dataset released! PDF: Code & Dataset: Full Narrated Video: High-speed flight is essential for time- and battery-constrained missions (e.g., inspection, exploration, search & rescue). However, fast motion corrupts visual data with severe motion blur and introduces drift/noise in visual-inertial odometry, making NeRF-based 3D reconstruction particularly brittle. We propose a unified framework that leverages asynchronous #EventCamera streams together with motion-blurred frames to reconstruct high-fidelity radiance fields from agile drone flights. Our key idea is to embed event-image fusion directly into radiance field optimization while jointly refining a shared, continuous-time camera trajectory initialized from event-based VIO. This enables us to recover sharp radiance fields and accurate trajectories without ground-truth supervision during training. We validate our method on synthetic data and on real sequences captured by a drone flying up to 2 m/s. Despite severe blur and noisy pose priors, our method preserves fine scene details and achieves a performance gain of over 50% on real-world data compared to state-of-the-art methods. Kudos to Rong Zou and Marco Cannici! Marco Cannici Reference: Rong Zou*, Marco Cannici*, Davide Scaramuzza Event-Aided Sharp Radiance Field Reconstruction for Fast-Flying Drones IEEE Transactions on Robotics (T-RO), 2026 NCCR Robotics European Research Council (ERC) AUTOASSESS UZH IfI University of Zurich UZH Science Prophesee SynSense UZH Space Hub

Davide Scaramuzza

12,006 views • 5 months ago

We are excited to share our #CORL2024 paper on learning quadrotor obstacle avoidance from the visual stream of a single #eventcamera! Trained entirely in simulation! We demonstrate obstacle avoidance both in the dark and in a forest up to 5m/s. PDF: Video: Project page: Event cameras are sensors that output per-pixel-level intensity changes at microsecond latency resolution; they feature nearly zero motion blur and high dynamic range but produce a very large volume of events under significant ego-motion and further lack a high-fidelity continuous-time sensor model in simulation, making direct #sim2real transfer not possible. By leveraging depth prediction as a pretext task, we pre-train a reactive obstacle avoidance policy with “approximated” simulated events and then fine-tune the perception component with limited events-and-depth real-world data. This technique bridges the sim2real gap for #eventcameras! As at the current state, there is no continuous-time sensor model for event cameras, we hope that this work can finally spur future research leveraging simulation for training event-vision-based policies to create faster, agile robots! Kudos to Anish Bhattacharya, @marcocannic, Vijay Kumar Nikolai Matni UZH Science University of Zurich UZH Space Hub UZH IfI European Research Council (ERC) GRASP Laboratory Penn Engineering
1:25

We are excited to share our #CORL2024 paper on learning quadrotor obstacle avoidance from the visual stream of a single #eventcamera! Trained entirely in simulation! We demonstrate obstacle avoidance both in the dark and in a forest up to 5m/s. PDF: Video: Project page: Event cameras are sensors that output per-pixel-level intensity changes at microsecond latency resolution; they feature nearly zero motion blur and high dynamic range but produce a very large volume of events under significant ego-motion and further lack a high-fidelity continuous-time sensor model in simulation, making direct #sim2real transfer not possible. By leveraging depth prediction as a pretext task, we pre-train a reactive obstacle avoidance policy with “approximated” simulated events and then fine-tune the perception component with limited events-and-depth real-world data. This technique bridges the sim2real gap for #eventcameras! As at the current state, there is no continuous-time sensor model for event cameras, we hope that this work can finally spur future research leveraging simulation for training event-vision-based policies to create faster, agile robots! Kudos to Anish Bhattacharya, @marcocannic, Vijay Kumar Nikolai Matni UZH Science University of Zurich UZH Space Hub UZH IfI European Research Council (ERC) GRASP Laboratory Penn Engineering

Davide Scaramuzza

17,219 views • 1 year ago

Glad that our work “Inference-Time Enhancement of Generative Robot Policies via Predictive World Modeling”, led by Han Qi, has been accepted to IEEE Robotics and Automation Letters! 🎉 We propose Generative Predictive Control (GPC): sample action proposals from a pretrained diffusion policy (“look back”), roll them out with a diffusion-based action-conditioned video world model (“look forward”), then rank or optimize the actions using either a learned reward model or VLM preferences. Conceptually, this is trajectory optimization / MPC with hybrid sampling + gradient optimization, interpreted through modern diffusion priors and video world models. Interestingly, we first posted the paper on arXiv in Feb 2025, when action-conditioned video world models for planning were still rare—now this direction is rapidly gaining traction. Still many open questions, e.g., • how to avoid local minima in planning • what representations work best for world models • how to balance physics priors vs. data-driven learning Paper:
1:00

Glad that our work “Inference-Time Enhancement of Generative Robot Policies via Predictive World Modeling”, led by Han Qi, has been accepted to IEEE Robotics and Automation Letters! 🎉 We propose Generative Predictive Control (GPC): sample action proposals from a pretrained diffusion policy (“look back”), roll them out with a diffusion-based action-conditioned video world model (“look forward”), then rank or optimize the actions using either a learned reward model or VLM preferences. Conceptually, this is trajectory optimization / MPC with hybrid sampling + gradient optimization, interpreted through modern diffusion priors and video world models. Interestingly, we first posted the paper on arXiv in Feb 2025, when action-conditioned video world models for planning were still rare—now this direction is rapidly gaining traction. Still many open questions, e.g., • how to avoid local minima in planning • what representations work best for world models • how to balance physics priors vs. data-driven learning Paper:

Heng Yang

18,994 views • 4 months ago

TokenFlow: Consistent Diffusion Features for Consistent Video Editing paper page: The generative AI revolution has recently expanded to videos. Nevertheless, current state-of-the-art video models are still lagging behind image models in terms of visual quality and user control over the generated content. In this work, we present a framework that harnesses the power of a text-to-image diffusion model for the task of text-driven video editing. Specifically, given a source video and a target text-prompt, our method generates a high-quality video that adheres to the target text, while preserving the spatial layout and motion of the input video. Our method is based on a key observation that consistency in the edited video can be obtained by enforcing consistency in the diffusion feature space. We achieve this by explicitly propagating diffusion features based on inter-frame correspondences, readily available in the model. Thus, our framework does not require any training or fine-tuning, and can work in conjunction with any off-the-shelf text-to-image editing method. We demonstrate state-of-the-art editing results on a variety of real-world videos.
0:36

TokenFlow: Consistent Diffusion Features for Consistent Video Editing paper page: The generative AI revolution has recently expanded to videos. Nevertheless, current state-of-the-art video models are still lagging behind image models in terms of visual quality and user control over the generated content. In this work, we present a framework that harnesses the power of a text-to-image diffusion model for the task of text-driven video editing. Specifically, given a source video and a target text-prompt, our method generates a high-quality video that adheres to the target text, while preserving the spatial layout and motion of the input video. Our method is based on a key observation that consistency in the edited video can be obtained by enforcing consistency in the diffusion feature space. We achieve this by explicitly propagating diffusion features based on inter-frame correspondences, readily available in the model. Thus, our framework does not require any training or fine-tuning, and can work in conjunction with any off-the-shelf text-to-image editing method. We demonstrate state-of-the-art editing results on a variety of real-world videos.

AK

143,469 views • 3 years ago

France Minister of Transportation on Tesla's FSD in new video: "The safety trade-offs are not yet sufficient to authorize it as it currently stands. All the more so since this system is not a self-driving system; So what are our concerns? Without going into overly technical detail, there are several. First example: we've found that the system allows for speeding. Another example: according to the data we have, the system doesn't guarantee an optimal level of driver attention in urban settings and during the most complex maneuvers, such as lane changes, intersections, or roundabouts. So what are the next steps? We're currently continuing our technical discussions with the Netherlands, other European states, and Tesla to refine our analysis. Last week, a representative from the Ministry of Transport's services traveled to the Netherlands. I want to emphasize this: France supports these technologies, and their deployment must happen within a framework of trust that allows us to evaluate the systems and determine what adjustments FSD needs to guarantee a high level of safety. This autumn, I will also be bringing together the French autonomous vehicle ecosystem to present the levers that will accelerate the deployment of this technology and to set a direction for the coming years. In France, we're fortunate to have world-class automakers and automotive suppliers, as well as some of the best engineers in the world in artificial intelligence and digital technology. We have real cards to play, and we're ready to play them — but never at the expense of safety on our roads."
4:57

France Minister of Transportation on Tesla's FSD in new video: "The safety trade-offs are not yet sufficient to authorize it as it currently stands. All the more so since this system is not a self-driving system; So what are our concerns? Without going into overly technical detail, there are several. First example: we've found that the system allows for speeding. Another example: according to the data we have, the system doesn't guarantee an optimal level of driver attention in urban settings and during the most complex maneuvers, such as lane changes, intersections, or roundabouts. So what are the next steps? We're currently continuing our technical discussions with the Netherlands, other European states, and Tesla to refine our analysis. Last week, a representative from the Ministry of Transport's services traveled to the Netherlands. I want to emphasize this: France supports these technologies, and their deployment must happen within a framework of trust that allows us to evaluate the systems and determine what adjustments FSD needs to guarantee a high level of safety. This autumn, I will also be bringing together the French autonomous vehicle ecosystem to present the levers that will accelerate the deployment of this technology and to set a direction for the coming years. In France, we're fortunate to have world-class automakers and automotive suppliers, as well as some of the best engineers in the world in artificial intelligence and digital technology. We have real cards to play, and we're ready to play them — but never at the expense of safety on our roads."

Sawyer Merritt

704,721 views • 16 days ago

Check out our #PAMI paper with code "Dense Continuous-Time Optical Flow from Event Cameras," where we show how to regress *continuous-time* trajectories of every pixel from event cameras alone or events plus frames! The key idea is to iteratively estimate per-pixel polynomials using a recurrent lookup and update scheme. Paper: Code: DOI: We present a method for estimating dense continuous-time optical flow from event data. Traditional dense optical flow methods compute the pixel displacement between two images. Due to missing information, these approaches cannot recover the pixel trajectories in the blind time between two images. We show that it is possible to compute per-pixel, continuous-time optical flow using events from an event camera. Events provide temporally fine-grained information about movement in pixel space due to their asynchronous nature and microsecond response time. We leverage these benefits to predict pixel trajectories densely in continuous time via parameterized Bézier curves. To achieve this, we build a neural network with strong inductive biases for this task: First, we build multiple sequential correlation volumes in time using event data. Second, we use Bézier curves to index these correlation volumes at multiple timestamps along the trajectory. Third, we use the retrieved correlation to update the Bézier curve representations iteratively. Our method can optionally include image pairs to boost performance further. To train and evaluate our model, we introduce a synthetic dataset (MultiFlow) that features moving objects and ground truth trajectories for every pixel. Our quantitative experiments suggest that our method successfully predicts pixel trajectories in continuous time and is competitive in the traditional two-view pixel displacement metric on MultiFlow and DSEC-Flow. Open source code and datasets are released to the public. Kudos to Mathias Gehrig Manasi Muglikar
0:23

Check out our #PAMI paper with code "Dense Continuous-Time Optical Flow from Event Cameras," where we show how to regress *continuous-time* trajectories of every pixel from event cameras alone or events plus frames! The key idea is to iteratively estimate per-pixel polynomials using a recurrent lookup and update scheme. Paper: Code: DOI: We present a method for estimating dense continuous-time optical flow from event data. Traditional dense optical flow methods compute the pixel displacement between two images. Due to missing information, these approaches cannot recover the pixel trajectories in the blind time between two images. We show that it is possible to compute per-pixel, continuous-time optical flow using events from an event camera. Events provide temporally fine-grained information about movement in pixel space due to their asynchronous nature and microsecond response time. We leverage these benefits to predict pixel trajectories densely in continuous time via parameterized Bézier curves. To achieve this, we build a neural network with strong inductive biases for this task: First, we build multiple sequential correlation volumes in time using event data. Second, we use Bézier curves to index these correlation volumes at multiple timestamps along the trajectory. Third, we use the retrieved correlation to update the Bézier curve representations iteratively. Our method can optionally include image pairs to boost performance further. To train and evaluate our model, we introduce a synthetic dataset (MultiFlow) that features moving objects and ground truth trajectories for every pixel. Our quantitative experiments suggest that our method successfully predicts pixel trajectories in continuous time and is competitive in the traditional two-view pixel displacement metric on MultiFlow and DSEC-Flow. Open source code and datasets are released to the public. Kudos to Mathias Gehrig Manasi Muglikar

Davide Scaramuzza

12,637 views • 2 years ago

Autonomous driving and obstacles avoidance at drift speeds, challenging the limits of what is possible! In this demo, our vehicle can be seen performing #autonomousdriving at very high speeds, causing it to both skid and drift at turns, while also avoiding obstacles. At such speeds, given the inherent dynamics of the vehicle platform used, it is very easy for the vehicle to topple. The #reinforcementlearning based motion planning and decision making framework that is being demoed here is tasked with ensuring obstacles avoidance without compromising on the speed, to an extent possible, and to drive the vehicle as fast as possible. This is evident towards the end of the video, where it can be seen that our vehicle avoided static obstacles while drifting.This demonstrates the level of sophistication and agility in our framework to ensure proper control of the #autonomousvehicles at high-speeds. The use cases are many; to begin with, our generic off-roads autonomous driving research focuses on enabling autonomous navigation in previously unknown and unseen environments, while ensuring mathematical completeness guarantees. Such agility can also help on-road autonomous vehicles to deal with unforeseeable corner cases or sudden appearance of obstacles in its tracks, at high-speeds. Our underlying research at Swaayatt Robots is still far from over, and over the next 3-4 months, we will be demonstrating abstract representation being learned by our multi-RL agents based framework (under progress) to ensure computation of the cost of the terrain without any labelled data, where multiple agents learn to control / regulate different aspects of autonomous navigation, to ensure safe and robust navigation, both on- and off-roads. All the people on the ground, who participated in the demo, were trained safety professionals. #deeplearning #MachineLearning #Robotics
3:38

Autonomous driving and obstacles avoidance at drift speeds, challenging the limits of what is possible! In this demo, our vehicle can be seen performing #autonomousdriving at very high speeds, causing it to both skid and drift at turns, while also avoiding obstacles. At such speeds, given the inherent dynamics of the vehicle platform used, it is very easy for the vehicle to topple. The #reinforcementlearning based motion planning and decision making framework that is being demoed here is tasked with ensuring obstacles avoidance without compromising on the speed, to an extent possible, and to drive the vehicle as fast as possible. This is evident towards the end of the video, where it can be seen that our vehicle avoided static obstacles while drifting.This demonstrates the level of sophistication and agility in our framework to ensure proper control of the #autonomousvehicles at high-speeds. The use cases are many; to begin with, our generic off-roads autonomous driving research focuses on enabling autonomous navigation in previously unknown and unseen environments, while ensuring mathematical completeness guarantees. Such agility can also help on-road autonomous vehicles to deal with unforeseeable corner cases or sudden appearance of obstacles in its tracks, at high-speeds. Our underlying research at Swaayatt Robots is still far from over, and over the next 3-4 months, we will be demonstrating abstract representation being learned by our multi-RL agents based framework (under progress) to ensure computation of the cost of the terrain without any labelled data, where multiple agents learn to control / regulate different aspects of autonomous navigation, to ensure safe and robust navigation, both on- and off-roads. All the people on the ground, who participated in the demo, were trained safety professionals. #deeplearning #MachineLearning #Robotics

Sanjeev Sharma

11,181 views • 1 year ago

InstantDrag Improving Interactivity in Drag-based Image Editing discuss: Drag-based image editing has recently gained popularity for its interactivity and precision. However, despite the ability of text-to-image models to generate samples within a second, drag editing still lags behind due to the challenge of accurately reflecting user interaction while maintaining image content. Some existing approaches rely on computationally intensive per-image optimization or intricate guidance-based methods, requiring additional inputs such as masks for movable regions and text prompts, thereby compromising the interactivity of the editing process. We introduce InstantDrag, an optimization-free pipeline that enhances interactivity and speed, requiring only an image and a drag instruction as input. InstantDrag consists of two carefully designed networks: a drag-conditioned optical flow generator (FlowGen) and an optical flow-conditioned diffusion model (FlowDiffusion). InstantDrag learns motion dynamics for drag-based image editing in real-world video datasets by decomposing the task into motion generation and motion-conditioned image generation. We demonstrate InstantDrag's capability to perform fast, photo-realistic edits without masks or text prompts through experiments on facial video datasets and general scenes. These results highlight the efficiency of our approach in handling drag-based image editing, making it a promising solution for interactive, real-time applications.
0:46

InstantDrag Improving Interactivity in Drag-based Image Editing discuss: Drag-based image editing has recently gained popularity for its interactivity and precision. However, despite the ability of text-to-image models to generate samples within a second, drag editing still lags behind due to the challenge of accurately reflecting user interaction while maintaining image content. Some existing approaches rely on computationally intensive per-image optimization or intricate guidance-based methods, requiring additional inputs such as masks for movable regions and text prompts, thereby compromising the interactivity of the editing process. We introduce InstantDrag, an optimization-free pipeline that enhances interactivity and speed, requiring only an image and a drag instruction as input. InstantDrag consists of two carefully designed networks: a drag-conditioned optical flow generator (FlowGen) and an optical flow-conditioned diffusion model (FlowDiffusion). InstantDrag learns motion dynamics for drag-based image editing in real-world video datasets by decomposing the task into motion generation and motion-conditioned image generation. We demonstrate InstantDrag's capability to perform fast, photo-realistic edits without masks or text prompts through experiments on facial video datasets and general scenes. These results highlight the efficiency of our approach in handling drag-based image editing, making it a promising solution for interactive, real-time applications.

AK

71,232 views • 1 year ago

Announcing DreamDojo: our open-source, interactive world model that takes robot motor controls and generates the future in pixels. No engine, no meshes, no hand-authored dynamics. It's Simulation 2.0. Time for robotics to take the bitter lesson pill. Real-world robot learning is bottlenecked by time, wear, safety, and resets. If we want Physical AI to move at pretraining speed, we need a simulator that adapts to pretraining scale with as little human engineering as possible. Our key insights: (1) human egocentric videos are a scalable source of first-person physics; (2) latent actions make them "robot-readable" across different hardware; (3) real-time inference unlocks live teleop, policy eval, and test-time planning *inside* a dream. We pre-train on 44K hours of human videos: cheap, abundant, and collected with zero robot-in-the-loop. Humans have already explored the combinatorics: we grasp, pour, fold, assemble, fail, retry—across cluttered scenes, shifting viewpoints, changing light, and hour-long task chains—at a scale no robot fleet could match. The missing piece: these videos have no action labels. So we introduce latent actions: a unified representation inferred directly from videos that captures "what changed between world states" without knowing the underlying hardware. This lets us train on any first-person video as if it came with motor commands attached. As a result, DreamDojo generalizes zero-shot to objects and environments never seen in any robot training set, because humans saw them first. Next, we post-train onto each robot to fit its specific hardware. Think of it as separating "how the world looks and behaves" from "how this particular robot actuates." The base model follows the general physical rules, then "snaps onto" the robot's unique mechanics. It's kind of like loading a new character and scene assets into Unreal Engine, but done through gradient descent and generalizes far beyond the post-training dataset. A world simulator is only useful if it runs fast enough to close the loop. We train a real-time version of DreamDojo that runs at 10 FPS, stable for over a minute of continuous rollout. This unlocks exciting possibilities: - Live teleoperation *inside* a dream. Connect a VR controller, stream actions into DreamDojo, and teleop a virtual robot in real time. We demo this on Unitree G1 with a PICO headset and one RTX 5090. - Policy evaluation. You can benchmark a policy checkpoint in DreamDojo instead of the real world. The simulated success rates strongly correlate with real-world results - accurate enough to rank checkpoints without burning a single motor. - Model-based planning. Sample multiple action proposals → simulate them all in parallel → pick the best future. Gains +17% real-world success out of the box on a fruit packing task. We open-source everything!! Weights, code, post-training dataset, eval set, and whitepaper with tons of details to reproduce. DreamDojo is based on NVIDIA Cosmos, which is open-weight too. 2026 is the year of World Models for physical AI. We want you to build with us. Happy scaling! Links in thread:
1:17

Announcing DreamDojo: our open-source, interactive world model that takes robot motor controls and generates the future in pixels. No engine, no meshes, no hand-authored dynamics. It's Simulation 2.0. Time for robotics to take the bitter lesson pill. Real-world robot learning is bottlenecked by time, wear, safety, and resets. If we want Physical AI to move at pretraining speed, we need a simulator that adapts to pretraining scale with as little human engineering as possible. Our key insights: (1) human egocentric videos are a scalable source of first-person physics; (2) latent actions make them "robot-readable" across different hardware; (3) real-time inference unlocks live teleop, policy eval, and test-time planning *inside* a dream. We pre-train on 44K hours of human videos: cheap, abundant, and collected with zero robot-in-the-loop. Humans have already explored the combinatorics: we grasp, pour, fold, assemble, fail, retry—across cluttered scenes, shifting viewpoints, changing light, and hour-long task chains—at a scale no robot fleet could match. The missing piece: these videos have no action labels. So we introduce latent actions: a unified representation inferred directly from videos that captures "what changed between world states" without knowing the underlying hardware. This lets us train on any first-person video as if it came with motor commands attached. As a result, DreamDojo generalizes zero-shot to objects and environments never seen in any robot training set, because humans saw them first. Next, we post-train onto each robot to fit its specific hardware. Think of it as separating "how the world looks and behaves" from "how this particular robot actuates." The base model follows the general physical rules, then "snaps onto" the robot's unique mechanics. It's kind of like loading a new character and scene assets into Unreal Engine, but done through gradient descent and generalizes far beyond the post-training dataset. A world simulator is only useful if it runs fast enough to close the loop. We train a real-time version of DreamDojo that runs at 10 FPS, stable for over a minute of continuous rollout. This unlocks exciting possibilities: - Live teleoperation *inside* a dream. Connect a VR controller, stream actions into DreamDojo, and teleop a virtual robot in real time. We demo this on Unitree G1 with a PICO headset and one RTX 5090. - Policy evaluation. You can benchmark a policy checkpoint in DreamDojo instead of the real world. The simulated success rates strongly correlate with real-world results - accurate enough to rank checkpoints without burning a single motor. - Model-based planning. Sample multiple action proposals → simulate them all in parallel → pick the best future. Gains +17% real-world success out of the box on a fruit packing task. We open-source everything!! Weights, code, post-training dataset, eval set, and whitepaper with tons of details to reproduce. DreamDojo is based on NVIDIA Cosmos, which is open-weight too. 2026 is the year of World Models for physical AI. We want you to build with us. Happy scaling! Links in thread:

Jim Fan

226,180 views • 5 months ago

You might have seen the WuBOT performing at the 2026 Spring Festival Gala; however, most high-dynamic extreme motions you see are executed by overfitted tracking policies. Until now, training a unified policy capable of performing various extreme motions with a high success rate remained an unsolved challenge. We spent an entire year digging into the barrier between general tracking and extreme physical behaviors. After burning through dozens of G1 robots, we finally identified the bottleneck of learning and physical executability. With these discoveries, we developed OmniXtreme: the first general policy that can execute diverse extreme motions, including consecutive flips, extreme balancing, and even breakdancing with rapid contact switches! This capability is achieved by pre-training a flow-based generative control policy and then post-training with actuation-aware residual RL for complex physical dynamics—a step we found critical for successful real-world transfer. This work is a joint collaboration with Unitree. Together, we are pushing the physical limits of humanoid robots. It is incredibly exciting to see a general "robot gymnast" and "robot breakdancer" come to life! It was also our first time publishing a paper with XingXing, which was an enlightening experience. The model checkpoints are now released—we welcome you to play with them! 📦 📄 Paper: 🌐 Project: 💻 Code:
1:20

You might have seen the WuBOT performing at the 2026 Spring Festival Gala; however, most high-dynamic extreme motions you see are executed by overfitted tracking policies. Until now, training a unified policy capable of performing various extreme motions with a high success rate remained an unsolved challenge. We spent an entire year digging into the barrier between general tracking and extreme physical behaviors. After burning through dozens of G1 robots, we finally identified the bottleneck of learning and physical executability. With these discoveries, we developed OmniXtreme: the first general policy that can execute diverse extreme motions, including consecutive flips, extreme balancing, and even breakdancing with rapid contact switches! This capability is achieved by pre-training a flow-based generative control policy and then post-training with actuation-aware residual RL for complex physical dynamics—a step we found critical for successful real-world transfer. This work is a joint collaboration with Unitree. Together, we are pushing the physical limits of humanoid robots. It is incredibly exciting to see a general "robot gymnast" and "robot breakdancer" come to life! It was also our first time publishing a paper with XingXing, which was an enlightening experience. The model checkpoints are now released—we welcome you to play with them! 📦 📄 Paper: 🌐 Project: 💻 Code:

Siyuan Huang

107,262 views • 5 months ago