Dense temporal motion retargeting improves dynamic robot motions
Dense Temporal Motion Retargeting for Legged Robots
Robotics
Summary
Legged robots can copy movements from humans or animals, but because their bodies are different, the motions must be carefully adjusted. The authors introduced a method called Dense Temporal Motion Retargeting (DTMR) that changes the timing and control of robot movements at every small step to make movements more natural and closer to the original. This approach works especially well for complicated actions like jumps and runs. They showed DTMR is faster and more accurate than other methods and that it can be used on real humanoid robots.
What this means in practice
- •For legged robot engineers: Enable better adaptation of human and animal motions to legged robots, improving dynamic movements like jumping and running.
- •For animation software developers: Provide faster and more precise tools for retiming motions in humanoid character animations, enhancing realism and control.
Authors
Jaeryeong Kim, Taerim Yoon, Jin Cheng, Sungjoon Choi, Stelian Coros
Abstract
Legged robots can learn expressive whole-body skills from the motions of humans and animals. Due to the morphology gap between the source and the robot, however, the motion must be tailored to the dynamic properties of the robot. In particular, dynamic motions such as a jump require careful adjustment, since their timing and control are interdependent. We propose dense temporal motion retargeting (DTMR), which jointly optimizes timing and control within a single optimization, where dense means that the timing is adjusted for every control step. This dense formulation enables DTMR to deform only the parts of the motion that need a change in timing. The problem is solved with sampling-based model predictive control (MPC) in parallel on a GPU. We evaluate DTMR against baselines on two hours of human motion with four humanoid robots, where the results show that DTMR outperforms baseline methods, particularly on dynamic motions. We also show that allowing more temporal deformation yields more precise retargeting. We further compare DTMR with a baseline that optimizes the temporal dimension, where the result shows that DTMR retargets more precisely under the same deformation budget while being ~19x faster. Lastly, policies trained on our references transfer to a real humanoid robot.