🤖 AI Summary
To address the challenges of multi-behavior coordinated learning, smooth mode transitions, and cross-task generalization in quadrupedal robots, this paper proposes the Versatile Instructable Motion prior (VIM) framework. Methodologically, VIM introduces a novel dual-objective reward design—comprising functional and stylistic rewards—and unifies motion imitation, sparse instruction guidance, kinematic constraints, and sim-to-real transfer within a single end-to-end policy. This enables concurrent learning and execution of over ten high-dynamic, low-level locomotion skills—including running, turning, jumping, and backflips. Our key contributions are: (i) the first demonstration of real-time learning and seamless switching among diverse agile locomotion behaviors using a single controller; (ii) substantial improvements in task coverage and motion naturalness; and (iii) simultaneous validation in both simulation and on physical hardware, with demonstrated cross-domain policy transfer capability.
📝 Abstract
The agility of animals, particularly in complex activities such as running, turning, jumping, and backflipping, stands as an exemplar for robotic system design. Transferring this suite of behaviors to legged robotic systems introduces essential inquiries: How can a robot learn multiple locomotion behaviors simultaneously? How can the robot execute these tasks with a smooth transition? How to integrate these skills for wide-range applications? This paper introduces the Versatile Instructable Motion prior (VIM) - a Reinforcement Learning framework designed to incorporate a range of agile locomotion tasks suitable for advanced robotic applications. Our framework enables legged robots to learn diverse agile low-level skills by imitating animal motions and manually designed motions. Our Functionality reward guides the robot's ability to adopt varied skills, and our Stylization reward ensures that robot motions align with reference motions. Our evaluations of the VIM framework span both simulation and the real world. Our framework allows a robot to concurrently learn diverse agile locomotion skills using a single learning-based controller in the real world. Videos can be found on our website: https://rchalyang.github.io/VIM/
