Humanoid Robot

Motivation:

    By the mid 21 century, a team of fully autonomous humanoid robot soccer player will defeat human soccer players. This is one aim of the development of humanoid robot. Humanoid robot is defined as a robot with human-like form and behavior features, which can communicate with people and perform tasks in human living conditions by using sensors systems such as vision and auditory. Compared with other robots, humanoid have some advantages in motion and communication. On the one hand, humanoid robots are able to move in areas that are normally inaccessible to wheeled robots, such as stairs and areas littered with obstacles. On the other hand, it may be easier for people to interact with robots with a humanoid shape rather than robots with a nonhuman shape. As an important topic of robot, humanoid can make good contribution to the people's lives. Humanoid robot not only operate as workers in areas that are dangerous to humans, but also work as domestic assistants or nurses because of their humanlike shape.

Approach:

    The aim of this project is to design and build a real humanoid robot with human behaviors that is capable of walking, turning, crouching, avoiding obstacles, running and communicating with peoples. Based on the sensors information, the robot can make compensation to keep a stable motion.

    The research includes three sub-directions:

    1. motion control, including high performance mobility ,behavioral robustness, behavioral complexity.

    2. sensory systems design, including vision, auditory and force sensors, sensor fusion.

    3. system design, including actuation selection and mechanical design.

People:

Chenggang Liu is exploring motion control and skill learning, responsibilities include system architecture, motion control system design and implementation, biped locomotion, and full-body motion control.

Dengpeng Xing is exploring motion control algorithm, responsibilities include mechanical design, optimization, and simulations.

Yi Zhang is responsible for control system hardware design and implementation.

Breen Yang is responsible for control system software design and implementation.

Former menbers:

Xusheng Lie, Jian Chen, Fei Liu

Publications:

  1. Chenggang Liu and Jianbo Su, Basic Behavior Acquisition Based on Multisensor Integration of a Robot Head, IEEE International Conference on Robotics and Automation, 2008, accepted
  2. Chenggang Liu and Jianbo Su, Gaze Stabilization Acquisition Based on Multi-sensor Integration, The 7th World Congress on Intelligent Control and Automation, 2008, submitted
  3. Xusheng Lei, Jianbo SU, ¡°Application of RBFNN for Humanoid Robot Real Time Optimal Trajectory Generation in Running¡±, In Lecture Notes in Computer Science (Advances in Neural Networks¡ª Proceedings of International Symposium on Neural Networks), Vol. 3174, pp. 1-6, Dalian, China, Aug., 2004.
  4. Xusheng Lei, Jianbo SU, ¡°Running Trajectory Generation for Humanoid Robot¡±, IEEE International Conference on Machine Learning and Cybernetics, Shanghai, China, Aug. 2004.
  5. Jianbo SU, ¡°Optimal Incremental Approach to the Motion Compression for Telepresent Locomotion¡±, Proceedings of IEEE Inter. Conf. On Robotics and Automation, pp.1664-1669, New Orleans, USA, April, 2004.
  6. Yanjun Zhang, Jianbo SU, ¡°Prototype Design of the Plug-and-Play Desktop Robotic System¡±, IEEE International Conference on Machine Learning and Cybernetics, Shanghai, China, Aug. 2004.
  7. Jing Pan, Jianbo Su, Xusheng Lei, Feng Lan, ¡°Motion Control Strategy for Humanoid Robot MIH-1¡±, Proceedings of the IEEE International Conference on Mechatronics & Automation, pp819-824, Canada, July 2005
  8. Xusheng Lei, Jianbo Su, Feedback Control of Humanoid Robot Locomotion, LNAI 3613, uzzy Systems and Knowledge Discovery-FSKD 2005,pp. 890-899, Changsha, China, Aug.,2005.
  9. Xusheng Lei, Jing Pan, Jianbo Su, Humanoid Robot Locomotion, IEEE International Conference on Machine Learning and Cybernetics, pp.882-887, Guangzhou, china, 2005.

Simulations:

Photo Gallery: