Path Planning Algorithm Research Of Robot

Path planning technology as the key of robot field, now, has been deeply researched andwidely applied, and also its core is the path planning algorithm. Path planning refers: in anenvironment with obstacles, according to a certain evaluation standard, finds an optimal pathwithout collision from the starting point to the target point. The paper mainly researches onrobot path planning algorithms. The paper firstly introduces the path planning basic principle,several commonly environment models, some commonly used programming algorithms, andsummarizes the advantages and disadvantages of the algorithms. The RRT algorithm hasrandomness, also does not need planning environment modeling in the planning process. Italso has a probabilistic completeness, as long as the environment has an effective path, thealgorithm can obtain the feasible path. Therefore it has been widely applied in robot pathplanning. This paper mainly studies the RRT algorithm, and then presents two improvedalgorithms. The details are as follows:For the basic RRT algorithm having the problems of large randomness, slowconvergence speed and low search efficiency, the variable probability guiding RRT algorithmis presented in this paper. Based on the bias-goal RRT algorithm, by setting the variableprobability parameters and local guiding point, makes the exploring random tree quickly growin the direction of the target point; and by setting the variable step function, makes the randomtree effectively avoid obstacles, and accelerates the path convergence speed.In order to realize the real-time path planning of mobile robot under dynamic uncertainenvironment, this paper proposes an improved algorithm combined with artificial potentialfield and variable probability guiding RRT algorithm. Firstly, in the initial environment withfewer obstacles, artificial potential field method is used for quickly pre-planning initial path.When a new obstacle appears and covers the initial path, the real-time planning is started.Determining appropriate planning space around the new obstacle, a local path is planned by variable probability guiding RRT algorithm. At last, the final planning path is obtainedthrough merging the initial path and local path.Different simulation environments are set, and MATLAB is used for simulationexperiments. Two improved algorithms are used for simulation experiments in differentenvironments, determining the selection of parameters and also validating the improvedalgorithms feasibility and effectiveness. Simulation results show that the improved algorithmsare feasible. Compared with other algorithms, the improved algorithms have good pathoptimization and obstacle avoidance capability.