Publications

Manipulate as Human: Learning Task-oriented Manipulation Skills by Adversarial Motion Priors

Published in Submitted to IEEE RAL and Under Review, 2024

In recent years, there has been growing interest in developing robots and autonomous systems that can interact with humans in a more natural and intuitive way. One of the key challenges in achieving this goal is to enable these systems to manipulate objects and tools in a manner that is similar to how humans do. In this paper, we propose a novel approach for learning human-style manipulation skills by using adversarial motion priors.

Recommended citation: Ziqi Ma, Changda Tian and Yue Gao. Manipulate as Human: Learning Task-oriented Manipulation Skills by Adversarial Motion Priors. Submitted to IEEE ICRA 2024 and Under Review, Sept 2023.

Learning Capability to Enhance Locomotion Control and Planning for Legged Robots

Published in Submitted to IEEE RAM and Under Review, 2023

The capability of a legged robot varies in line with its structure, topology, and locomotion controller. With different capability, legged robots can walk past terrains of varying complexity. However, existing legged robots’ motion controllers and path planners neither consider how to maximize the robot’s capability according to the current environment, nor give a global guidance path that meets the robot’s capability. In this letter, we propose a hierarchical learning based motion control and planning system for legged robots. It consists of three parts: capability maximizing, capability abstraction and capability based path planning. Capability maximizing uses reinforcement learning to choose best control strategy and robot topology which maximizes the robot’s traverse capability; capability abstraction network gives the robot’s capability under current terrain and footholds state; capability based path planning gives long-range guidance path that conforms our robots traverse capability learned by capability abstraction network. We trained our framework in simulation and did long-range locomotion experiments both in simulation and real world on our hexapod robot Qingzhui. The results shows that our method greatly improves the global locomotion performance of our legged robot.

Recommended citation: Changda Tian and Yue Gao. Learning Capability to Enhance Locomotion Control and Planning for Legged Robots. Submitted to IEEE RAM and under review, Sept 2023.

CapPlanner: Adaptable to Various Topology and Locomotion Capability for Hexapod Robots

Published in ROBIO2022, Best Paper in Biomimetics Finalist, 2022

We present CapPlanner, a hierarchical motion control and planning system which can do long-range locomotion control and planning according to the learned traverse capability of the robot in different topologies. It consists of two layers, the bottom-level controller computes the trajectory of the body and the feet according to the terrain, local target and current feets’ positions. Besides, it controls the motors to track the calculated trajectory. The top-level controller learns the traverse ability of the robot with its bottom-level controller by simulating locomotion tasks on various terrains and in different topologies. Hence our CapPlanner can guide the robot to reach a long-term destination with a much higher success rate.

Recommended citation: Changda Tian and Yue Gao. CapPlanner: Adaptable to Various Topology and Locomotion Capability for Hexapod Robots. In IEEE International Conference on Robotics and Biomimetics (IEEE-ROBIO), December 2022.

Sim-to-Real: Designing Locomotion Controller for Six-Legged Robot

Published in IEEE International Conference on CYBER Technology in Automation, Control, and Intelligent Systems (IEEE-CYBER), 2019

Use sim-to-real reinforcement learning to train a hexpod learn to walk to its target at the same time get rid of the obstacles.

Recommended citation: Chenyu Yang, Changda Tian, Qingshan Yao and Yue Gao. Sim-to-Real: Designing Locomotion Controller for Six-Legged Robot. In IEEE International Conference on CYBER Technology in Automation, Control, and Intelligent Systems (IEEE-CYBER), July 2019.