Video yükleniyor...

Video Yüklenemedi

Ana Sayfaya Dön

releasing AME2: Agile and Generalized Legged Locomotion via Attention-Based Neural Map Encoding In this work, we discuss how to achieve a combination of generalization and agility in legged locomotion, and propose a general solution.

ChongZzZhang's profile picture

C. Zhang

6,949 subscribers

92,748 görüntüleme • 5 ay önce •via X (Twitter)

EğitimBilim & Teknoloji
LIVE

Anya Rossi• Live Now

Private livecam show

0 Yorum

Yorum bulunmuyor

Orijinal gönderinin yorumları burada görünecek

Benzer Videolar

The teaser video for our latest Science Robotics paper. Full video will be out soon. Paper: Attention-based map encoding for learning generalized legged locomotion Link:
0:54

The teaser video for our latest Science Robotics paper. Full video will be out soon. Paper: Attention-based map encoding for learning generalized legged locomotion Link:

Junzhe (JJ) He

13,162 görüntüleme • 10 ay önce

How to break locomotion agile-safe tradeoffs?Introduce Agile But Safe: Learning Collision-Free High-Speed Legged Locomotion: - Fully onboard - Agile (>3m/s) - Safe (collision-free guarantee) - Robust & versatile How? RL + model-free reach-avoid value! 👉
1:30

How to break locomotion agile-safe tradeoffs?Introduce Agile But Safe: Learning Collision-Free High-Speed Legged Locomotion: - Fully onboard - Agile (>3m/s) - Safe (collision-free guarantee) - Robust & versatile How? RL + model-free reach-avoid value! 👉

Tairan He

128,469 görüntüleme • 2 yıl önce

AGILE Foundation Model Redefines Integrated Perception & Locomotion AGIBOT introduces AGILE — the AGIBOT Generative Intelligent Locomotion Engine — its latest integrated perception-locomotion foundation model. Empowered by visual awareness, AGIBOT X2 achieves seamless integration of perception and locomotion. When robots truly perceive the world, locomotion becomes more intelligent. #AGIBOT #Humanoid #AGILE #EmbodiedAI #PhysicalAI
3:03

AGILE Foundation Model Redefines Integrated Perception & Locomotion AGIBOT introduces AGILE — the AGIBOT Generative Intelligent Locomotion Engine — its latest integrated perception-locomotion foundation model. Empowered by visual awareness, AGIBOT X2 achieves seamless integration of perception and locomotion. When robots truly perceive the world, locomotion becomes more intelligent. #AGIBOT #Humanoid #AGILE #EmbodiedAI #PhysicalAI

AGIBOT

24,457 görüntüleme • 1 ay önce

Even after 4yrs of locomotion research, we keep getting surprised by how far we can push the limits of legged robots! We report a major update 🚀🤖 Extreme Parkour: extremely long & high jumps, ramp, handstand, etc. all with a single neural net! 🧵(1/n)
1:36

Even after 4yrs of locomotion research, we keep getting surprised by how far we can push the limits of legged robots! We report a major update 🚀🤖 Extreme Parkour: extremely long & high jumps, ramp, handstand, etc. all with a single neural net! 🧵(1/n)

Deepak Pathak

260,141 görüntüleme • 2 yıl önce

AGIBOT has unveiled AGILE (AgiBot Generative Intelligent Locomotion Engine), a perception-control foundation model for whole-body humanoid locomotion. - It fuses visual perception, balance, and motion planning end-to-end, replacing the traditional split between "seeing" and "moving." - The robot reads terrain and obstacles in real time and adjusts its gait with no preset trajectories. - It runs on local compute with millisecond-level response and works across AGIBOT's lineup (A1/A2/A3, X2). AGILE is the locomotion "cerebellum" that complements AGIBOT's Genie Operator "brain."
0:35

AGIBOT has unveiled AGILE (AgiBot Generative Intelligent Locomotion Engine), a perception-control foundation model for whole-body humanoid locomotion. - It fuses visual perception, balance, and motion planning end-to-end, replacing the traditional split between "seeing" and "moving." - The robot reads terrain and obstacles in real time and adjusts its gait with no preset trajectories. - It runs on local compute with millisecond-level response and works across AGIBOT's lineup (A1/A2/A3, X2). AGILE is the locomotion "cerebellum" that complements AGIBOT's Genie Operator "brain."

The Humanoid Hub

14,020 görüntüleme • 1 ay önce

I appreciate the confidence by Rainbow Robotics when they drove the robot dog into crowds. #CoRL2025 Very swift and agile. I rarely call a locomotion controller elegant but this is how it felt. Long video in 🧵
0:22

I appreciate the confidence by Rainbow Robotics when they drove the robot dog into crowds. #CoRL2025 Very swift and agile. I rarely call a locomotion controller elegant but this is how it felt. Long video in 🧵

Chenhao Li

69,836 görüntüleme • 9 ay önce

enjoyed hosting CNET at Foundation Robotics SF office. We discuss defense usecases, teleop, RL locomotion and more
7:29

enjoyed hosting CNET at Foundation Robotics SF office. We discuss defense usecases, teleop, RL locomotion and more

Sankaet

113,210 görüntüleme • 9 ay önce

Truly general-purpose robots must be able to navigate spaces they have never seen before. ⦿ Skild Brain enables end-to-end autonomous locomotion from raw vision and joint inputs, without mapping or pre-planning. ⦿ The model adapts in real time to new terrain such as stairs, gaps, and clutter, reacting instinctively like a human. There is no “stair mode” or “step-over mode”: the transition between walking behaviors is continuous and instinctive. ⦿ A single neural policy supports a wide range of locomotion tasks and can transfer across different robot platforms.
4:37

Truly general-purpose robots must be able to navigate spaces they have never seen before. ⦿ Skild Brain enables end-to-end autonomous locomotion from raw vision and joint inputs, without mapping or pre-planning. ⦿ The model adapts in real time to new terrain such as stairs, gaps, and clutter, reacting instinctively like a human. There is no “stair mode” or “step-over mode”: the transition between walking behaviors is continuous and instinctive. ⦿ A single neural policy supports a wide range of locomotion tasks and can transfer across different robot platforms.

The Humanoid Hub

28,439 görüntüleme • 11 ay önce

This robodog aims to remove humans from hazardous work environments The Robot Teleoperativo 2 project out of the Dynamic Legged Systems lab has developed an advanced teleoperation system designed to operate in high-risk environments, minimizing dangers to human workers. By integrating cutting-edge technologies in tele-locomotion, tele-manipulation, and remote human-robot interaction, the system enables precise, dual-arm control for complex tasks in hazardous conditions.
1:01

This robodog aims to remove humans from hazardous work environments The Robot Teleoperativo 2 project out of the Dynamic Legged Systems lab has developed an advanced teleoperation system designed to operate in high-risk environments, minimizing dangers to human workers. By integrating cutting-edge technologies in tele-locomotion, tele-manipulation, and remote human-robot interaction, the system enables precise, dual-arm control for complex tasks in hazardous conditions.

Circuit

22,827 görüntüleme • 1 yıl önce

1/N Excited to share the lab’s first preprint on neural population dynamics in motor cortex and spinal motoneurons during locomotion! This project was led by Eric Kirk, an outstanding postdoc in the lab. Thread …
0:30

1/N Excited to share the lab’s first preprint on neural population dynamics in motor cortex and spinal motoneurons during locomotion! This project was led by Eric Kirk, an outstanding postdoc in the lab. Thread …

Britton Sauerbrei

68,433 görüntüleme • 2 yıl önce

Quadrupeds are fast. Agile. Great at locomotion. But can they manipulate? A new approach from Carnegie Mellon University, Google DeepMind, and Bosch is teaching quadrupedal robots to do more than walk, they’re learning to interact. It’s called Human2LocoMan: a system that uses human data to pretrain robot policies before finetuning on real hardware. The result? A four-legged robot that can walk, carry, organize, scoop, and sort with both single and dual-arm control. By pretraining on human motion, they cut the amount of robot data in half—while improving success rates by over 80% in unfamiliar environments. Their Modularized Cross-Embodiment Transformer (MXT) learns from both human and robot demonstrations, then generalizes those skills to physical tasks—no hardcoded behaviors required. It’s locomotion and manipulation. A quadruped that can walk and clean up after itself?
1:21

Quadrupeds are fast. Agile. Great at locomotion. But can they manipulate? A new approach from Carnegie Mellon University, Google DeepMind, and Bosch is teaching quadrupedal robots to do more than walk, they’re learning to interact. It’s called Human2LocoMan: a system that uses human data to pretrain robot policies before finetuning on real hardware. The result? A four-legged robot that can walk, carry, organize, scoop, and sort with both single and dual-arm control. By pretraining on human motion, they cut the amount of robot data in half—while improving success rates by over 80% in unfamiliar environments. Their Modularized Cross-Embodiment Transformer (MXT) learns from both human and robot demonstrations, then generalizes those skills to physical tasks—no hardcoded behaviors required. It’s locomotion and manipulation. A quadruped that can walk and clean up after itself?

Lukas Ziegler

60,546 görüntüleme • 1 yıl önce

One-legged and one-armed woman with wheelchair spotted twerking in front of a Harlem bodega
0:34

Sensitive content

One-legged and one-armed woman with wheelchair spotted twerking in front of a Harlem bodega

Daily Loud

850,106 görüntüleme • 11 ay önce

Motion Matching based locomotion system in #UE5 (Gameplay ready & Fully Replicated)
2:19

Motion Matching based locomotion system in #UE5 (Gameplay ready & Fully Replicated)

Kelechi Apakama

28,161 görüntüleme • 2 yıl önce

Multi legged robots built for tough terrain! 🐛 Unlike wheeled or long-legged machines that stall in cluttered fields, their multi legged design provides stable, low-profile locomotion through confined and unstructured environments. The robots from Ground Control Robotics move close to the ground, giving direct access to crops while carrying precision sensors, from hyperspectral cameras to soil moisture and VOC (Volatile Organic Compound) detectors, without disturbing fragile gases. Applications range from agriculture to defense, search & rescue, and pest control, but the team is first targeting the multi-billion dollar specialty agriculture market, where traditional robots fail to operate effectively. Developed in collaboration with Georgia Institute of Technology, the platform is positioned as a versatile sensor and mobility system for environments that defeat conventional designs.
0:52

Multi legged robots built for tough terrain! 🐛 Unlike wheeled or long-legged machines that stall in cluttered fields, their multi legged design provides stable, low-profile locomotion through confined and unstructured environments. The robots from Ground Control Robotics move close to the ground, giving direct access to crops while carrying precision sensors, from hyperspectral cameras to soil moisture and VOC (Volatile Organic Compound) detectors, without disturbing fragile gases. Applications range from agriculture to defense, search & rescue, and pest control, but the team is first targeting the multi-billion dollar specialty agriculture market, where traditional robots fail to operate effectively. Developed in collaboration with Georgia Institute of Technology, the platform is positioned as a versatile sensor and mobility system for environments that defeat conventional designs.

Lukas Ziegler

44,288 görüntüleme • 10 ay önce

I won’t stop talking about Locomotion until EVERY fitness professional in the world teaches Perfectly Balanced Locomotion. So simple. And relatively easy (as perfectly balanced locomotion is more natural than walking and running out of balance - name an animal or ancient human that lacks perfect balance). #HeadOverFoot We need standards in the fitness industry. To NOT begin with balanced gait is…. You don’t know how. To upset balance and uproot people with stiff spine restrictive tension as the foundation is…. What peer review exercise did, and continues to do. And I believe it’s a primary reason we’ve seen epidemic proportions of Non-Contact Injuries. 100% True. #EveryStepStronger
1:45

I won’t stop talking about Locomotion until EVERY fitness professional in the world teaches Perfectly Balanced Locomotion. So simple. And relatively easy (as perfectly balanced locomotion is more natural than walking and running out of balance - name an animal or ancient human that lacks perfect balance). #HeadOverFoot We need standards in the fitness industry. To NOT begin with balanced gait is…. You don’t know how. To upset balance and uproot people with stiff spine restrictive tension as the foundation is…. What peer review exercise did, and continues to do. And I believe it’s a primary reason we’ve seen epidemic proportions of Non-Contact Injuries. 100% True. #EveryStepStronger

David Weck

53,066 görüntüleme • 1 yıl önce

I'm starting a new series of tutorials that explain the creature's locomotion in detail. We'll go through each part of the state machine and discuss various tricks and non-obvious moments. The rig and animations are free and available for download. All links below:
0:45

I'm starting a new series of tutorials that explain the creature's locomotion in detail. We'll go through each part of the state machine and discuss various tricks and non-obvious moments. The rig and animations are free and available for download. All links below:

Slava Borovik

51,791 görüntüleme • 11 ay önce

Learning H-Infinity Locomotion Control Stable locomotion in precipitous environments is an essential capability of quadruped robots, demanding the ability to resist various external disturbances. However, recent learning-based policies only use basic domain randomization to
2:01

Learning H-Infinity Locomotion Control Stable locomotion in precipitous environments is an essential capability of quadruped robots, demanding the ability to resist various external disturbances. However, recent learning-based policies only use basic domain randomization to

AK

38,639 görüntüleme • 2 yıl önce

See Spot perform dynamic whole-body manipulation. Using a combination of reinforcement learning (RL) and sampling-based control, the robot is able to autonomously drag, roll, and stack tires weighing 15 kg (33 lb), well above its maximum arm lift capacity. Learn more about coordinating locomotion and manipulation processes:
0:27

See Spot perform dynamic whole-body manipulation. Using a combination of reinforcement learning (RL) and sampling-based control, the robot is able to autonomously drag, roll, and stack tires weighing 15 kg (33 lb), well above its maximum arm lift capacity. Learn more about coordinating locomotion and manipulation processes:

RAI Institute

87,489 görüntüleme • 9 ay önce