| In recent years,with the development of artificial intelligence and the needs of society,indoor service robot technology is also in production,research and innovative progress.In order to improve the autonomy of the robot,it is necessary to enhance its navigation planning ability,so that it can ensure its own safety in the case of various scenes,and achieve sufficient perception of environment and obstacle avoidance.The main research contents of this thesis include the construction of a layered costmaps based on the fusion of multiple sensor data,the introduction of a human leg detection model to solve the problem of moving pedestrians during navigation,and the improvement of traditional navigation strategies and recovery mechanisms.First of all,in view of the defects existing in the current navigation process,the thesis introduces a forward planning strategy to enhance the robustness of local obstacle avoidance.Moreover,the thesis introduces a centered planning strategy to ensure the safety of the robot when passing through a narrow area.And the strategies are called according to the distance measured by sensors and the time of the path planning In order to prevent the robot navigation from being blocked,the thesis improves and updates the existing recovery mechanism.The escape strategy is mainly introduced to help the robot escape from the obstacles in a timely manner,and the costmaps is reset and updated through the map recovery strategy to ensure that the robot can successfully complete the navigation.Furthermore,the self-built robot platform is used for experimental verification.It mainly compares and analyzes the planning path between the improved planning strategy and the traditional planning strategy;and carries out relevant verification on the improved recovery mechanism.Test the effect of solving the "pedestrian problem"after introducing the human leg detection model;test the safety and success rate of the robot when navigating indoors using the navigation strategy in this paper.The experimental results prove that by combining the improved planning strategy and the"pedestrian problem" removal method,it can effectively achieve obstacle avoidance for mobile pedestrians and static tables and chairs,and has good application prospects.Eventually,the self-built robot platform is used for experimental verification.It mainly compares and analyzes the planning path between the improved planning strategy and the traditional planning strategy.Then,experiment has been conducted to verify the improved recovery mechanism.It tests the effect of solving the "pedestrian problem" after introducing the human leg detection model.Meanwhile,it tests the safety and success rate of the robot when navigating indoors using the navigation strategy in this thesis.The experiment proves that by combining the improved planning strategy and the "pedestrian problem" removal method,robots can effectively avoid colliding with pedestrians and fixed tables and chairs.It wisnesses a bright future of application. |
