Research On Geometric Modeling And Autonomous Exploration Strategies Of Mobile Robot For Underground Environment

The mobile robot has been used more and more in the production and life of human beings because of their good endurance and strong carrying capacity.Mobile robots with autonomous environment exploration capabilities are mostly used in extreme scenarios such as military reconnaissance,accident site exploration,and scientific exploration.This research is based on the Cross-Dangerous Underground Challenge Competition,with a crawler mobile robot as the research object,to carry out research on autonomous exploration strategies for the underground environment,and realize autonomous exploration in small and medium-scale underground environments through terrain analysis algorithms and autonomous environment exploration strategies.First of all,for the underground environment without GPS signal,poor lighting conditions and wireless signal shielding,an unmanned underground environment exploration system is built.On the basis of existing crawler-type mobile platform,the wheel-crawler-leg composite mobile mechanism is designed to adapt to the unstructured terrain;The fusion module of 3D lidar and IMU sensor is designed to solve the problem of no GPS signal and poor illumination conditions;The remote communication framework is built,setting up a local area network to solve the signal shielding problem,and finally form a set of autonomous exploration unmanned platform system for the underground environment.Secondly,aiming at the low efficiency of robot autonomous environment exploration due to many obstacles and narrow spaces in the underground environment,the terrain analysis algorithm is designed based on the elevation map.The general grid map library is used to construct a local elevation map that follows the movement of robot.According to the information in the elevation map and the robot's ability to overcome obstacles,the robot's accessibility to each grid position in the elevation map is analyzed.In the form of a two-dimensional occupancy grid map,three kinds of passable maps with different size ranges are constructed for the robot.Thirdly,in order to realize the efficient environmental exploration for the robot in the underground environment,an autonomous environment exploration strategy is proposed based on the random sampling algorithm and the improved graph search algorithm.The strategy includes:exploring boundary point selection,random node sampling,geometric topology map construction,exploration path planning,topology path optimization and tracking processes.Random node sampling is performed based on the RRT algorithm,the Dijkstra algorithm is improved for the exploring path planning,and the maximum likelihood estimation method is used to achieve the robot's path tracking.Finally,in order to verify the effort of the autonomous exploration strategy for the underground environment,a simulation experiment system and a physical experiment system are built respectively,and the experimental scheme is designed.Four indicators are proposed to evaluate the performance of the autonomous environment exploration strategy,including the length of trajectory,the time-consuming of autonomous exploration,the coverage of exploration space and the speed change of robot.The experimental results prove that the autonomous exploration strategy can efficiently complete the autonomous exploration task with an exploration coverage rate of more than 80%and an exploration rate of more than 10m~3/s in the underground environment.