Path Planning For Mobile Robots Based On Digital Elevation Model Grid Map

Mobile robot navigation is the key technology of the study of mobile robots. Path Planning is an important element of mobile robots navigation, which provides the information of robots to the motion control during moving. The task of path planning can be described as follows:Find a collision-free path in the environment which has barriers according to given evaluation standard. This paper researches deeply the path planning technology for mobile robots in the complex environment. The following aspects are investigated and discussed:1) A traversability analysis method based on Digital Elevation Model grid is presented. And the results of the analysis are classified to generate a cost map. With the extracted basic terrain information based on Digital Elevation Model of environment, traversability of the terrain is analyzed, comprehensively considering the factors such as mechanical properties and kinematic constraints of mobile robots. Then get the final cost map according to the classified terrain features.2) Aimed at some problems caused by the traditional path planning algorithms when they are used in the complex environment. A motion constrained generalized Field D* algorithm is proposed. In this algorithm, the cost function is designed to take into account one or several of the navigation costs, including distance, safety and time cost. Moreover, the planned path is further smoothed regarding to the constraint of minimum turning radius for a robot. The smoothed path is therefore easier for the robot to follow. A group of simulated grid maps describing complex environments have been tested. The experiments show that the proposed algorithm not only fits complicated environments but also improves the performability of the results.3) On the basis of the analysis of the advantages and disadvantages of current global and local path planning algorithms, a hierarchical path planning approach is given. This approach combines the typical local obstacle avoidance algorithm Morphin with global path planning algorithm Field D*, executing with the organic integration of global information and local information. The approach not only can optimize the motion path for the robot globally, but also avoid local extrema like many traditional local planning methods. The simulation results show that the approach has good adaptability and strong robustness.