Vision-based Leader-follower Formation Control Of Mobile Robots In Uncalibrated Environments

One of the research focuses in cooperation of multi-robot is formation control. The formation control based on a more convenient and low-cost method not only depends on smart mechanism but also relies on advanced control algorithm. Therefore, the development of the adaptive image-based Leader-Follower formation control algorithm is necessary.This paper studies at the problem of the image-based Leader-Follower formation control of mobile robots using an uncalibrated on-board perspective camera or an uncalibrated on-board omnidirectional camera independent of both the measurement of the height of the feature point on Leader and the communication between them. The unknown intrinsic and extrinsic parameters of the perspective camera or the unknown extrinsic parameters of the omnidirectional camera as well as the unknown coefficient of the motion plane of the feature point are all parameterize linearly depending on the assumption of planar motion of the mobile robots. Adaptive control theory is utilized to design the control algorithm. Additionally, a filter witch can generate an image-based signal to eliminate the demand of the measurement of the Leader’s velocity. The Lyapunov method proves the stability for the system. The simulation and the experiments validate the proposed adaptive control algorithm.