Research On Path Planning Of Mobile Robots

Path planning for robots is a key research area in robotics, also is the hotspot of the certain research domain. This paper is based on realistic demand in science projects and takes deep dig into motion planning in static environment and path planning in dynamic environment.Robots system in static environment.The final proposed solutions not only aimed at the shortcomings of the ant colony algorithm in robot path planning,but also in dynamic path planning unknown environment is put forward a new local path planning method. In traditional path algorithm problems in to the known and unknown environment path planning are subject to improvement and innovation, and simulation experiment, Experimental results show the new algorithm in reliability and validity, good performance.Main contents of the paper are as follows:1. Aimed at the shortcomings of the ant colony algorithm in robot path planning, which need much time and easy to fall in premature stagnation. This paper proposes a random expansion ant colony optimization algorithm through giving a possible way in the initial pheromone distribution to narrow the searching range of algorithm and raise the searching speed. At the same time randomly expansion factor is introduced to ant colony optimization algorithm, improved the diversity of path and global optimization properties. The simulations result shows that the algorithm has excellent global optimization property and fast convergence speed.2. Aiming at the robot dynamic path planning problem, this paper proposed a new method which implemented real-time path planning of mobile robot in complex dynamic environment. This method is based on path planning and obstacle avoidance strategy for rolling windows, according to visible point sub-goal setting, obstacle detour, analysis and forecast for moving obstacle, realized robot path planning in complex dynamic environment. Aiming at obstacles distribution, reasonable exchange between visible point and circumvent algorithm based on the obstacle distribution state, this could resolve infinite loop and minima that happened in local path planning. This proposed method suitable for dynamic environment that both convex and concave obstacle exist, simulation result proved its validity and feasibility.