| Since the 20th century,with the rapid rise of the field of artificial intelligence,artificial intelligence technology has begun to be widely used in the field of robotics and intelligent robots have emerged.As the most important category of intelligent robotics,mobile robots have always focused on the attention of academics at home and abroad.As the key point of mobile robots,how the robots "move",that is,how to perform path planning,has naturally become the most critical step in the study of mobile robots.Aiming at the path planning problem in the working environment with obstacles in different states,this paper proposes a potential field-ant colony fusion algorithm to guide the mobile robot to plan the path.On this basis,further research on the path planning problem of the mobile robot under complex terrain conditions.The main content of the work is as follows:(1)Traditional ant colony algorithm has strong robustness and high planning efficiency,so its application frequency in the field of path planning is relatively high.But at the same time,defects such as poor globality and slow convergence speed also make the path planned by the ant colony algorithm often non-optimal.Aiming at such problems,this paper introduces the target point information of the artificial potential field method into the heuristic function,which makes the target point continue to play a role in the whole planning process,and enhances the overall nature of the algorithm.At the same time,the artificial potential field force information is introduced into the heuristic function,so that the ant colony can rely on this information to escape and continue searching when it falls into a local minimum.The maximum-minimum ant system idea is introduced to limit the range of pheromone,which can prevent If the element changes too much,the convergence speed is slow or too fast,and a local optimal solution appears.(2)For the path planning problem of mobile robots with global and local environments,a hierarchical path planning method is proposed.In path planning,according to the motion state of obstacles,the path planning is divided into global and local levels.When planning at the global level,the potential field-ant colony algorithm is used to plan the path.In addition,in order to enhance the ability of the proposed algorithm to escape the local minimum,a measure to shield U-shaped traps is proposed,and multiple U-shaped trap obstacles are set in the working environment to test the performance of the proposed method.In the local layer,when the mobile robot encounters a moving obstacle,the rolling window theory is used to detect dynamic obstacles in the environment,and the artificial potential field method is called for local obstacle avoidance.(3)For the optimal path output after the path planning is completed,the cubic B-spline algorithm is introduced for optimization processing,which can effectively smooth the path peaks and reduce the energy loss caused by the robot's unsmooth walking during the movement.(4)Focusing on the problem of excessive energy consumption of mobile robots caused by not considering terrain factors in traditional path planning,by assuming that the working environment of mobile robots is not completely flat terrain,a path planning method for mobile robots under multiple terrain constraints is proposed.In order to represent the uneven terrain environment,a 2.5-dimensional grid map is defined,the concepts of semi-free grid and semi-obstacle grid are introduced,and the original range of passability coefficients are expanded for mobile robot identification.Then,the total cost function is designed by integrating the three non-flat terrain factors of ground elevation,ground slope,and surface roughness,and corresponding to the passability coefficient,so as to judge whether the grid area is feasible.(5)An experimental environment was built in the MATLAB simulation platform and the experiment verified the effectiveness and feasibility of the proposed method. |
PDF Full Text Request