Research Of Multi-objective Mobile Robot Path Planning Algorithm Based On ROS

Robotics covers a wide range of disciplines and technologies,including mechanical manufacturing technology,sensor application and identification technology,electronic technology,automation technology and artificial intelligence technology.In recent years,the continuous development of automation and artificial intelligence technology has greatly promoted the progress of robot technology.According to different application scenarios,robots can be divided into industrial robots,agricultural robots,services robots,and medical robots.With the wide application and development of robots,the requirements for robot intellectualization are getting higher and higher.Among them,autonomous navigation is the key metrics to evaluate the intellectualization of a robot,and path planning is an important part of autonomous navigation.So path planning technology is becoming a key technology in recent years.The problem of path planning can be divided into single-objective path planning problem and multi-objective path planning problem.The single-objective path planning problem generally only considers a single optimization goal,which is generally path length or path safety.Compared with the problem of single-objective path planning,multi-objective path planning considers the length,security,smoothness and other factors at the same time,making the planned path more suitable for practical application scenarios.Robot path planning generally includes global path planning and local path planning.Aiming at these two problems,this paper proposes adaptive genetic algorithm and improved brainstorm algorithm respectively to solve the problem of global and local path planning.The global path planning uses the artificial potential field method to build the maps,and then uses the proposed adaptive operator and the supervise operator to optimize the path length,path safety and path smoothness that need to be considered in the global path planning.Finally a Pareto solution set is obtained,which including a set of feasible solutions for the global path planning problem.Finally,the optimal solution of the optimization is obtained according to the requirements.According to the requirement of optimization speed and real-time performance,local path planning improves the brainstorm algorithm.Aiming at the problem that the grouping strategy and updating method of the original brainstorm algorithm are prone to produce local optimal solution and the optimization iteration speed is slow,the improved strategy is proposed.By evenly grouping the solution space of the problem,the optimization group is distributed in the whole solution space,which solves the problem easily falling into local optimal.By selecting the updated individual,the problem of slow iteration speed when updating all individuals is solved.Finally,a brainstorming algorithm suitable for local path planning is obtained.Finally,the simulation robot was built based on ROS,and the camera and laser radar were used as sensors to obtain environmental information.The proposed path planning algorithms were used for path planning,and the feasibility of the algorithm was verified through path planning experiments on grid maps and ROS maps.