Loading video...

Video Failed to Load

Go Home

Mobile manipulator playing badminton using model-free RL policy

simonkalouche's profile picture

Simon Kalouche

10,479 subscribers

91,835 views • 1 year ago •via X (Twitter)

Science & TechnologyGamingEducation
LIVE

Anya Rossi• Live Now

Private livecam show

5 Comments

Simon Kalouche's profile picture
Simon Kalouche1 year ago

Paper:

The Humanoid Hub's profile picture
The Humanoid Hub1 year ago

So impressive that it can do this with just one camera. DeepMind’s table tennis robot needed 22 cameras for ball and paddle tracking.

David Hansen 🇺🇸🇳🇿's profile picture
David Hansen 🇺🇸🇳🇿1 year ago

This is awesome

Cheeku Mani Tripathi's profile picture
Cheeku Mani Tripathi1 year ago

someone who won’t blame the racquet for losing the point. just the reward function. 🏸🤖

π's profile picture
π1 year ago

Nice

Related Videos

PNDBotics’ humanoid robot Adam tackles difficult terrain using an RL-based control policy.
1:00

PNDBotics’ humanoid robot Adam tackles difficult terrain using an RL-based control policy.

The Humanoid Hub

26,656 views • 1 year ago

TidyBot++ is an open-source mobile manipulator optimized for household tasks. The robot can be teleoperated using a mobile phone interface, enabling data collection for imitation learning.
0:22

TidyBot++ is an open-source mobile manipulator optimized for household tasks. The robot can be teleoperated using a mobile phone interface, enabling data collection for imitation learning.

The Humanoid Hub

29,188 views • 1 year ago

👇Introducing DPPO, Diffusion Policy Policy Optimization DPPO optimizes pre-trained Diffusion Policy using policy gradient from RL, showing 𝘀𝘂𝗿𝗽𝗿𝗶𝘀𝗶𝗻𝗴 𝗶𝗺𝗽𝗿𝗼𝘃𝗲𝗺𝗲𝗻𝘁𝘀 over a variety of baselines across benchmarks and sim2real transfer
0:20

👇Introducing DPPO, Diffusion Policy Policy Optimization DPPO optimizes pre-trained Diffusion Policy using policy gradient from RL, showing 𝘀𝘂𝗿𝗽𝗿𝗶𝘀𝗶𝗻𝗴 𝗶𝗺𝗽𝗿𝗼𝘃𝗲𝗺𝗲𝗻𝘁𝘀 over a variety of baselines across benchmarks and sim2real transfer

Allen Ren

78,227 views • 1 year ago

playing badminton with vibri
0:43

playing badminton with vibri

P.Carredo

47,757 views • 2 months ago

✨ Meet YOR: Open-Source Bimanual Mobile Manipulator from New York University Fully open-source mobile manipulator with dual 6-DoF PiPER arms by AgileX Robotics, BOM cost only ~$10k! 🌐 #Robotics #OpenSource #AgileXRobotics #PiPER #NYU
0:27

✨ Meet YOR: Open-Source Bimanual Mobile Manipulator from New York University Fully open-source mobile manipulator with dual 6-DoF PiPER arms by AgileX Robotics, BOM cost only ~$10k! 🌐 #Robotics #OpenSource #AgileXRobotics #PiPER #NYU

AgileX Robotics

16,251 views • 3 months ago

cutie ningning playing badminton 🫶🏻
0:32

cutie ningning playing badminton 🫶🏻

🌸

30,356 views • 10 months ago

We are moving from model controller to RL for locomotion. The policy is starting to look quite robust. Even more convinced now that legs are easier than wheels, given the software complexity has shrunk tremendously with RL.
0:26

We are moving from model controller to RL for locomotion. The policy is starting to look quite robust. Even more convinced now that legs are easier than wheels, given the software complexity has shrunk tremendously with RL.

Sankaet

133,130 views • 9 months ago

🚨 Kanpur-based startup xTerra Robotics has unveiled Scorp, India's 1st legged mobile manipulator.
4:12

🚨 Kanpur-based startup xTerra Robotics has unveiled Scorp, India's 1st legged mobile manipulator.

Indian Tech & Infra

111,021 views • 5 months ago

Open Duck Mini is standing up using a policy learned with RL in simulation! We are still pushing locomotion, so if you have some knowledge of Isaac Gym and want to help feel free to join us! Thomas Wolf
0:20

Open Duck Mini is standing up using a policy learned with RL in simulation! We are still pushing locomotion, so if you have some knowledge of Isaac Gym and want to help feel free to join us! Thomas Wolf

Antoine Pirrone

66,051 views • 1 year ago

Thanks AK! Finally, robot can do continuous, agile, autonomous, adaptive jumping over stair and stepping stone Key idea: combine the pros of model-free RL and model-based control. RL (for CoM refs) + QP (for GRF) + WBC (for torque) Open-sourced:
0:10

Thanks AK! Finally, robot can do continuous, agile, autonomous, adaptive jumping over stair and stepping stone Key idea: combine the pros of model-free RL and model-based control. RL (for CoM refs) + QP (for GRF) + WBC (for torque) Open-sourced:

Guanya Shi

32,155 views • 1 year ago

Don’t forget to split, merge and optimise using our mobile app. It’s FREE
0:47

Don’t forget to split, merge and optimise using our mobile app. It’s FREE

PuntrrAi

156,078 views • 1 month ago

Using our brain simulator, we’ve trained a reinforcement learning agent to maximize bits per second. Here is the RL policy converting brain data to cursor control in simulation:
0:29

Using our brain simulator, we’ve trained a reinforcement learning agent to maximize bits per second. Here is the RL policy converting brain data to cursor control in simulation:

Neuralink

18,410 views • 1 year ago

Excited to share Flow Matching Policy Gradients: expressive RL policies trained from rewards using flow matching. It’s an easy, drop-in replacement for Gaussian PPO on control tasks.
0:29

Excited to share Flow Matching Policy Gradients: expressive RL policies trained from rewards using flow matching. It’s an easy, drop-in replacement for Gaussian PPO on control tasks.

David McAllister

150,884 views • 11 months ago

Badminton training using equipment 🏸🤖💪🔥😃
0:24

Badminton training using equipment 🏸🤖💪🔥😃

Axel

72,606 views • 9 months ago

Whole-body Large Behavior Model diffusion policy from Boston Dynamics and TRI. Most advanced mobile-manipulation demo with error recovery I've seen to date.
0:22

Whole-body Large Behavior Model diffusion policy from Boston Dynamics and TRI. Most advanced mobile-manipulation demo with error recovery I've seen to date.

Simon Kalouche

41,706 views • 10 months ago

Madam President playing badminton with Saina Nehwal.. what a beautiful video.. ♥️
0:23

Madam President playing badminton with Saina Nehwal.. what a beautiful video.. ♥️

Mr Sinha

1,202,408 views • 1 year ago

Yesterday, a resident shared this video from Alaknanda Apartments. Parks r well lit & makes them feel safer. Younger lot is enjoying playing on synthetic badminton court. All using MLA funds.
0:36

Yesterday, a resident shared this video from Alaknanda Apartments. Parks r well lit & makes them feel safer. Younger lot is enjoying playing on synthetic badminton court. All using MLA funds.

Saurabh Bharadwaj

50,056 views • 1 year ago

A husband is playing badminton with his pregnant wife 🏸❤️🤰🏻
0:31

A husband is playing badminton with his pregnant wife 🏸❤️🤰🏻

Spidey 🕸️

52,778 views • 1 month ago

Mobile-TeleVision follows our previous idea on upper-body and lower-body separation for control: 1⃣We train RL for lower-body locomotion and standing; 2⃣We use IK and retargeting directly for upper-body for high-precision and smooth manipulation. We avoid using RL for upper-body motion tracking. This is because fundamentally it is hard to avoid errors with RL tracking, and sim2real gap brings another gap. At this point, we still find using IK is more practical and precise.
1:15

Mobile-TeleVision follows our previous idea on upper-body and lower-body separation for control: 1⃣We train RL for lower-body locomotion and standing; 2⃣We use IK and retargeting directly for upper-body for high-precision and smooth manipulation. We avoid using RL for upper-body motion tracking. This is because fundamentally it is hard to avoid errors with RL tracking, and sim2real gap brings another gap. At this point, we still find using IK is more practical and precise.

Xiaolong Wang

32,662 views • 1 year ago

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 views • 2 years ago