Загрузка видео...

Не удалось загрузить видео

На главную

Autonomous drone racing demo with direct Pixel to Action and Deep RL. Robotics: Science and Systems

realyunlong's profile picture

Yunlong Song

1,243 subscribers

19,724 просмотров • 2 лет назад •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 0

Нет доступных комментариев

Здесь появятся комментарии из оригинального поста

Похожие видео

Scientists have developed SUPER, an autonomous drone that can navigate unknown and cluttered environments safely, which could help future search and rescue efforts. Learn more in Science Robotics:
1:05

Scientists have developed SUPER, an autonomous drone that can navigate unknown and cluttered environments safely, which could help future search and rescue efforts. Learn more in Science Robotics:

Science Magazine

46,125 просмотров • 1 год назад

$ONDS gives me exposure to the autonomy layer of drones through drone-in-a-box systems, autonomous networks, counter-drone technology and command-and-control software. Ondas could become a broader autonomous defense platform across aerial drones, counter-UAS, ground robotics, ISR, border security and other mission-critical use cases.
0:59

$ONDS gives me exposure to the autonomy layer of drones through drone-in-a-box systems, autonomous networks, counter-drone technology and command-and-control software. Ondas could become a broader autonomous defense platform across aerial drones, counter-UAS, ground robotics, ISR, border security and other mission-critical use cases.

Shay Boloor

468,500 просмотров • 2 месяцев назад

Our work, "A Primer on SO(3) Action Representations in Deep Reinforcement Learning," was accepted to #ICLR2026! We provide a systematic study of action representation choices in RL, showing that they fundamentally impact training stability and performance. #Robotics #AI #RL
0:21

Our work, "A Primer on SO(3) Action Representations in Deep Reinforcement Learning," was accepted to #ICLR2026! We provide a systematic study of action representation choices in RL, showing that they fundamentally impact training stability and performance. #Robotics #AI #RL

Learning Systems and Robotics Lab (is hiring!)

49,655 просмотров • 5 месяцев назад

🤖 OrionX Robotics is actively engaged in a fundraising round to accelerate growth, scale operations, and advance the development of next-generation autonomous robotics systems. We look forward to partnering with investors who share our vision for the future of intelligent robotics.
3:30

🤖 OrionX Robotics is actively engaged in a fundraising round to accelerate growth, scale operations, and advance the development of next-generation autonomous robotics systems. We look forward to partnering with investors who share our vision for the future of intelligent robotics.

OrionX Robotics

12,345 просмотров • 2 месяцев назад

AI in Motion: Autonomous Driving and Real-World Systems Live from NVIDIA GTC 2026, our correspondent Robert Scoble Robert Scoble is on the expo floor where AI is moving from theory into real-world applications—spanning robotics, AI infrastructure, and autonomous driving. The trend is clear: AI is rapidly expanding into robotics, intelligent manufacturing, and autonomous systems. Watch our Day 2 field report and stay with us as we continue exploring the future of AI. #AI #GTC2026 #Robotics #AutonomousDriving #NVIDIA
20:11

AI in Motion: Autonomous Driving and Real-World Systems Live from NVIDIA GTC 2026, our correspondent Robert Scoble Robert Scoble is on the expo floor where AI is moving from theory into real-world applications—spanning robotics, AI infrastructure, and autonomous driving. The trend is clear: AI is rapidly expanding into robotics, intelligent manufacturing, and autonomous systems. Watch our Day 2 field report and stay with us as we continue exploring the future of AI. #AI #GTC2026 #Robotics #AutonomousDriving #NVIDIA

China Asset Management

62,948 просмотров • 5 месяцев назад

Yesterday, we live demo-ed a “generalist” VLA for (I think) the first time ever to a broad audience Robotics: Science and Systems. Bring any object. Ask anything. New environment, new instructions, no fine-tuning. Just impeccable vibes! ✨
0:54

Yesterday, we live demo-ed a “generalist” VLA for (I think) the first time ever to a broad audience Robotics: Science and Systems. Bring any object. Ask anything. New environment, new instructions, no fine-tuning. Just impeccable vibes! ✨

Dhruv Shah

56,699 просмотров • 1 год назад

Allen Control Systems makes Bullfrog, an autonomous counter drone weapon station....and we are HIRING (roles at Walt Maciborski of @CBSAustin visited us to learn more about Bullfrog and how it meets the rising drone threat with unmatched precision:
2:32

Allen Control Systems makes Bullfrog, an autonomous counter drone weapon station....and we are HIRING (roles at Walt Maciborski of @CBSAustin visited us to learn more about Bullfrog and how it meets the rising drone threat with unmatched precision:

Steven Simoni

230,923 просмотров • 4 месяцев назад

Deployment-Ready RL: Pitfalls, Lessons, and Best Practices We’ve published a full transcript of a webinar with Kyle🤖🚀🦭 (UT Austin) and the Humanoid team on our blog. He explores Sim2Real RL: - Action Space - Observation Space - Dealing with Model Mismatch - Reward Tuning Intuition - RL with Motion References We’re sharing this to spread knowledge & help push humanoid robotics forward. Read or listen here:
1:35:16

Deployment-Ready RL: Pitfalls, Lessons, and Best Practices We’ve published a full transcript of a webinar with Kyle🤖🚀🦭 (UT Austin) and the Humanoid team on our blog. He explores Sim2Real RL: - Action Space - Observation Space - Dealing with Model Mismatch - Reward Tuning Intuition - RL with Motion References We’re sharing this to spread knowledge & help push humanoid robotics forward. Read or listen here:

Humanoid

60,003 просмотров • 11 месяцев назад

See the Iron Drone Raider™ in action. Our autonomous interceptor is built to protect critical infrastructure, population centers, and national airspace. Today’s European airport order underscores the growing global demand for layered, reliable counter-UAS defense. $ONDS Airobotics | American Robotics | Sentrycs
0:24

See the Iron Drone Raider™ in action. Our autonomous interceptor is built to protect critical infrastructure, population centers, and national airspace. Today’s European airport order underscores the growing global demand for layered, reliable counter-UAS defense. $ONDS Airobotics | American Robotics | Sentrycs

Ondas Inc.

60,611 просмотров • 9 месяцев назад

Demis Hassabis says Gemini 3 marks a watershed moment where systems are finally ready to become autonomous agents by the end of 2025, autonomous systems could handle full tasks, and within 18 months, we may see a robotics breakthrough "10 years of progress now happens every year"
1:21

Demis Hassabis says Gemini 3 marks a watershed moment where systems are finally ready to become autonomous agents by the end of 2025, autonomous systems could handle full tasks, and within 18 months, we may see a robotics breakthrough "10 years of progress now happens every year"

Haider.

37,373 просмотров • 6 месяцев назад

A small STM32 project can still teach serious robotics. This bionic butterfly toy uses an STM32F103 controller and Keil5 to reproduce flapping wing motion through embedded control and mechanical design. Projects like this are a practical entry point into biomimetic robotics lightweight actuators and autonomous flying systems.
0:39

A small STM32 project can still teach serious robotics. This bionic butterfly toy uses an STM32F103 controller and Keil5 to reproduce flapping wing motion through embedded control and mechanical design. Projects like this are a practical entry point into biomimetic robotics lightweight actuators and autonomous flying systems.

Techniahqrobot | humanoid robots

13,364 просмотров • 21 дней назад

Autonomous Counter-UAS 'Iron Drone Raider System' by American Robotics 🇺🇸. AI-driven tracking locks onto targets with real-time telemetry.
1:32

Autonomous Counter-UAS 'Iron Drone Raider System' by American Robotics 🇺🇸. AI-driven tracking locks onto targets with real-time telemetry.

DefenseTrends

36,565 просмотров • 6 месяцев назад

Soldiers + robotics = an unstoppable force. Currently, in its pilot phase, the Robotic Autonomous Systems Leader Tactics Course is integrating robotics into the DNA of the Infantry and Armor branches. 🎥 US Army Maneuver Center of Excellence
0:51

Soldiers + robotics = an unstoppable force. Currently, in its pilot phase, the Robotic Autonomous Systems Leader Tactics Course is integrating robotics into the DNA of the Infantry and Armor branches. 🎥 US Army Maneuver Center of Excellence

U.S. Army

36,442 просмотров • 5 месяцев назад

Announcing RoboFuME🤖💨, a system for autonomous & efficient real-world robot learning that 1. pre-trains a VLM reward model and a multi-task RL policy from diverse off-the-shelf demo data; 2. runs RL fine-tuning online with the VLM reward model. 🔗 🧵↓
0:21

Announcing RoboFuME🤖💨, a system for autonomous & efficient real-world robot learning that 1. pre-trains a VLM reward model and a multi-task RL policy from diverse off-the-shelf demo data; 2. runs RL fine-tuning online with the VLM reward model. 🔗 🧵↓

Jingyun Yang

88,294 просмотров • 2 лет назад

PC Gaming Show Tokyo Direct: Morbid Metal, an action and transformation roguelite. Free demo and early access this September 30.
1:11

PC Gaming Show Tokyo Direct: Morbid Metal, an action and transformation roguelite. Free demo and early access this September 30.

Pirat_Nation 🔴

32,507 просмотров • 10 месяцев назад

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

27,952 просмотров • 2 лет назад

Andy Lowery joins me for an outdoor demo of cutting-edge defense technology, showcasing the Leonidas autonomous ground vehicle and its role in modern battlefield protection. Built through a collaboration between Epirus, General Dynamics Land Systems, and Kodiak Defense, this mobile system integrates advanced radar, autonomous navigation, and directed energy capabilities to detect, track, and neutralize drone threats in real time. I need one of these for the studio. 😂 Epirus
12:59

Andy Lowery joins me for an outdoor demo of cutting-edge defense technology, showcasing the Leonidas autonomous ground vehicle and its role in modern battlefield protection. Built through a collaboration between Epirus, General Dynamics Land Systems, and Kodiak Defense, this mobile system integrates advanced radar, autonomous navigation, and directed energy capabilities to detect, track, and neutralize drone threats in real time. I need one of these for the studio. 😂 Epirus

Shawn Ryan

120,980 просмотров • 3 месяцев назад

At GEODNET, our miners earn $GEOD tokens by operating satellite reference stations right from their homes. 🏠 The data you contribute fuels precision in robotics, agriculture, autonomous systems, and the AR metaverse. Join us and make an impact with satellite data!
2:17

At GEODNET, our miners earn $GEOD tokens by operating satellite reference stations right from their homes. 🏠 The data you contribute fuels precision in robotics, agriculture, autonomous systems, and the AR metaverse. Join us and make an impact with satellite data!

GEODNET

28,031 просмотров • 1 год назад