Dissertation
Dissertation > Industrial Technology > Automation technology,computer technology > Automation technology and equipment > Robotics > Robot

Gait Planning of Planar Quasi-Passive Bipedal Robots

Author XieChaoPing
Tutor ZhangJianMing;XiongRong
School Zhejiang University
Course Pattern Recognition and Intelligent Systems
Keywords bipedal walking multi-body system TMT method gait planning
CLC TP242
Type Master's thesis
Year 2010
Downloads 73
Quotes 0
Download Dissertation

The humanoid robot is a kind of robot that is designed to imitate human motion and behavior. Since 1980’s, the related issues have became hot research topics in the robotics. Bipedal walking is the main characteristic of humanoid robots as distinguished from other robots, and it is also one of the most important research fields in the robotics. This paper studies the gait planning method for planar bipedal walker based on the research results at home and abroad. The researches include:1. The point mass model of the bipedal robot basing on the passive model is constructed, and TMT method is used to derive the kinetic equations of multi-body systems, such as adding active elements to the system, following the track of planned trajectory and dealing with the impaction. These equations lay the foundation for testifying the gait planning method.2. According to the walking condition of rimless wagon wheel, the feasibility of the planar bipedal walking is discussed. Robots will never fall backward as long as enough energy is provided to the center of the mass; if the swing leg moves fast enough, robots will never fall forward. Ankle actuation and trajectory tracking for swing leg are implemented, which create conditions for gait planning.3. For the ideal level ground, this paper proposes a gait planning method for walking at desired speed with fixed step length; for the level ground with obstacles or potholes, a gait planning method for walking at desired step length with fixed speed at the vertical position is proposed. The time for real-time computation is shortened by polling list. Due to the combination of trajectory tracking and passive walking, the energy efficiency is guaranteed and adaptability is enhanced.

Related Dissertations
More Dissertations