Study On The Optimal Obstacle/collision Avoidance Algorithm For Wheel Mobile Robots Based On Prediction Window

Obstacle avoidance,collision avoidance and obstacle avoidance of formation for wheeled mobile robots with autonomous perception,decision-making and executive ability are the important problems in the field of robot research,their basic goals are to control the wheeled mobile robots or their formation to navigate without collision and reach the specified destination in the environment with obstacles and other wheeled mobile robots,which has a wide application prospect in warehousing logistics,intelligent transportation,area search coverage and other tasks.However,the nonlinear,nonholonomic constraints,multi-inputs and multi-outputs characteristics of the kinematics model of wheeled mobile robot make its motion planning and control become a difficult problem in the field of robot research.Therefore,the research on obstacle avoidance,collision avoidance and obstacle avoidance of formation for wheeled mobile robots has important application value and academic significance.Based on the predictive window technology,the problems of obstacle avoidance,collision avoidance and obstacle avoidance of formation for wheeled mobile robots are studied in this paper.The main research works are as follows:Firstly,aiming at the collision-free motion problem of single wheeled mobile robot in the motion scene with obstacles,an optimal obstacle avoidance algorithm is proposed using the predictive window technology.According to the nonlinear characteristics of the kinematics model of wheeled mobile robot,a linear model is obtained by the input-output feedback linearizing.Based on the obtained linear model,all the relative acceleration changes that may cause the collision between robot and obstacle are defined as a set of acceleration change obstacle using prediction window technology.An optimal obstacle avoidance acceleration change is planned for wheeled mobile robot based on the acceleration change obstacle and converted to the actual collision-free control inputs.Matlab simulation verifies the effectiveness of the designed optimal obstacle avoidance algorithm,and the comparative simulation with existing literature shows that the designed optimal obstacle avoidance algorithm can make the change of the control inputs of robot more smooth and the range of change smaller in the process of obstacle avoidance.Then,aiming at the problem of collision avoidance between wheeled mobile robots,the proposed optimal obstacle avoidance algorithm is extended.In order to assign the collision avoidance responsibility to each wheeled mobile robot and realize the cooperative collision avoidance among wheeled mobile robots,a centralized cooperative optimal collision avoidance algorithm is designed.The relative acceleration change required for collision avoidance is assigned to each wheeled mobile robot through the central computing unit,and the cooperative collision avoidance of multiple wheeled mobile robots is realized.In order to improve the computational efficiency and robustness of cooperative optimal collision avoidance algorithm,the distributed cooperative optimal collision avoidance algorithm is designed.The control goal of distributed cooperative optimal collision avoidance is realized by transforming the constraint of acceleration change obstacle on the collision avoidance relative acceleration changes of two wheeled mobile robots into the constraint on collision-avoidance acceleration change of each robot.Aiming at the problem of collision avoidance between heterogeneous mobile robots,the above two cooperative optimal collision avoidance algorithms are extended and a new cooperative collision avoidance algorithm based on acceleration change obstacle and model predictive control is designed.Matlab simulation verifies the effectiveness of the designed optimal collision avoidance algorithm,and the comparative simulation with the existing literature shows that the designed optimal collision avoidance algorithm not only has advantages in the term of changes of control inputs,but also has higher real-time performance.Finally,aiming at the problem of collision-free motion of wheeled mobile robot formation in the motion scene with obstacles,an optimal obstacle avoidance algorithm for wheeled mobile robot formation is proposed based on predictive window.In order to facilitate the prediction of the collisions between robot formation and obstacles,the accelerations that could cause collision between robot formation and obstacle are defined as a set of formation acceleration obstacle using prediction window technology.Based on the formation acceleration obstacle,an optimal obstacle avoidance formation is selected for the robot formation from the formation template and the acceleration of motion is planned.The collision avoidance algorithm proposed in the last part is used to realize the collision-free transformation of robot formation.Matlab simulation verifies the effectiveness of the proposed optimal obstacle avoidance algorithm for robot formation.