This paper presents control method for multiple two-wheeled mobile robots moving in formation. Trajectory tracking algorithm from [7] is extended by collision avoidance, and is applied to the different type of formation task: each robot in the formation mimics motion of the virtual leader with a certain displacement. Each robot avoids collisions with other robots and circular shaped, static obstacles existing in the environment. Artificial potential functions are used to generate repulsive component of the control. Stability analysis of the closed-loop system is based on Lyapunov-like function. Effectiveness of the proposed algorithm is illustrated by simulation results.
The paper presents construction and control system of the climbing robot Safari designed at the Poznan University of Technology for inspection of high building walls, executed in order to evaluate their technical condition. Because such tasks are uncomfortable and very dangerous for humans, this mobile machine gives a possibility to observe and examine the state of the surface on which it is moving. The robot is a construction developed for walking on flat but uneven vertical and horizontal surfaces. Its on-board equipment provides ability to remotely examine and record images reflecting the robot’s surroundings. At the beginning of the paper, several concepts of existing climbing robots (four-legged, six-legged, sliding platform) are outlined. Next, the mechanical system of the Safari robot is presented with special emphasis on its kinematic equations and description of movement stages. Then, the on-board manipulator as well as the sensor and control systems are described.
We propose a class of m-crane control systems, that generalizes two- and three-dimensional crane systems. We prove that each representant of the described class is feedback equivalent to the second order chained form with drift. In consequence, we prove that it is differentially flat. Then we investigate its control properties and derive a control law for tracking control problem.