Reinforcement Learning for Versatile, Dynamic, and Robust Bipedal Locomotion Control
This work presents a comprehensive study on using deep reinforcement learning (RL) to create dynamic locomotion controllers for bipedal robots. Going beyond focusing on a single locomotion skill, we develop a general control solution that can be used for a range of dynamic bipedal skills, from periodic walking and running to aperiodic jumping and standing.
Our RL-based controller incorporates a novel dual-history architecture, utilizing both a long-term and short-term input/output (I/O) history of the robot. This control architecture, when trained through the proposed end-to-end RL approach, consistently outperforms other methods across a diverse range of skills in both simulation and the real world. The study also delves into the adaptivity and robustness introduced by the proposed RL system in developing locomotion controllers. We demonstrate that the proposed architecture can adapt to both time-invariant dynamics shifts and time-variant changes, such as contact events, by effectively using the robot’s I/O history. Additionally, we identify task randomization as another key source of robustness, fostering better task generalization and compliance to disturbances.
The resulting control policies can be successfully deployed on Cassie, a torque-controlled human-sized bipedal robot. This work pushes the limits of agility for bipedal robots through extensive real-world experiments. We demonstrate a diverse range of locomotion skills, including: robust standing, versatile walking, fast running with a demonstration of a 400-meter dash, and a diverse set of jumping skills, such as standing long jumps and high jumps.
Supplementary Videos:
- Running:
- 400m Dash:
- Walking:
1 view
158
47
4 weeks ago 00:02:02 12
DayZ Update Teaser
1 month ago 00:38:19 1
L’horreur existentielle de l’usine à trombones.
1 month ago 00:01:29 1
Introducing the World’s Coolest Humanoid Robot — EngineAI SE01!
1 month ago 00:25:48 1
What do tech pioneers think about the AI revolution? - BBC World Service
1 month ago 00:19:32 1
How to Make a Carbon Fiber Car Bonnet/Hood - Part 2/3 : Resin Infusion
1 month ago 00:01:44 1
Unitree Introducing | Unitree G1 Humanoid Agent | AI Avatar | Price from $16K
1 month ago 00:02:50 1
LimX Dynamics’ Biped Robot P1 Conquers the Wild Based on Reinforcement Learning
1 month ago 00:32:28 1
Free English Class! Topic: Our Daily Routines! 🐕⏰🥙 (Lesson Only)
1 month ago 00:11:18 1
Learn How To Talk About Your Daily Routine in English Part 2
1 month ago 00:07:45 1
Learn How To Talk About Your Daily Routine in English by Watching Me Act Out Mine
1 month ago 00:14:45 1
Как установить Stable Diffusion 3.5 Large и Turbo на компьютер? Пошаговая инструкция для Windows.
1 month ago 00:02:15 1
Number and Counting song | Learn Counting to 1000 | Math for 2nd Grade | Kids Academy
1 month ago 00:01:58 1
Morse Code Alphabet Receiving Practice (1)
1 month ago 00:09:12 3
Learn English Through Story Level 1, Graded Reader Level 1, Stories Short Beginners, Basic English
1 month ago 00:29:52 6
[LeatherCraft] Baguette Bag 4K / FREE PDF PATTERN
1 month ago 00:59:40 1
’Little Learning Machines’ Postmortem: A Game About Training Neural Networks
2 months ago 01:08:43 1
#dobetter Podcast Episode 4: Learned Behavior during Extinction
2 months ago 00:01:11 3
MEVIUS: A Quadruped Robot Easily Constructed through E-Commerce (Humanoids 2024)
2 months ago 00:11:46 1
Jobs and Occupations - Vocabulary for Kids - Compilation
2 months ago 00:02:28 1
Yamaha | Artist Profile | Krissy Morash of Escuela Grind
2 months ago 00:04:08 1
Wild animals for kids - Vocabulary for kids
2 months ago 00:10:12 1
Let’s Learn English Around the House and Home | English Video with Subtitles
2 months ago 00:32:16 1
The Teacher Series #10
2 months ago 00:03:30 1
The Ancient Egypt - 5 things you should know - History for kids