Research On Operation Control Of Unmanned Platform In Intelligent Workshop

To realize the intelligence of the workshop material transfer system and ensure the safe and reliable distribution of materials in the intelligent workshop,this paper takes the operation control of the unmanned platform of the intelligent workshop as the entry point,and conducts research on the unmanned platform multi-sensor fusion SLAM(Simultaneous Localization and Mapping),path planning and task scheduling.The main research contents of the thesis are as follows:(1)Aiming at the existing problems of the two-dimensional lidar and depth camera fusion SLAM,an improved SLAM method combining lidar and depth camera is proposed.First,the IMU(Inertial Measurement Unit)and the wheel odometer are integrated to obtain the preliminary pose of the unmanned platform based on the EKF(Extended Kalman Filter).Then the optimized pose and lidar map are obtained by Gmapping algorithm.And the depth camera map is constructed with optimized pose and depth point cloud.Finally,the lidar map and the depth camera map are merged to improve the accuracy and completeness.(2)To improve the adaptability of the unmanned platform in dynamic environment,a spatiotemporal grid map is constructed based on the traditional grid map,which can integrate static obstacles and dynamic obstacles in a unified manner.Then an improved spatiotemporal DWA(dynamic window approach)path planning algorithm is proposed,which can perform local path planning in the spatiotemporal grid map.Finally,the A*algorithm and the improved spatiotemporal DWA algorithm are fused,and a spatiotemporal grid map path planning method fused with the A* algorithm is proposed.Experiments show that this method can not only ensure the global optimality of the path,but also adapt to the dynamic environment.(3)According to the maximum task handling efficiency and the shortest task handling distance,a multi-unmanned platform and multi-task scheduling goal is established.A task scheduling model is established based on constraints,assumptions and task scheduling objectives,and a task allocation strategy based on dynamic priority is proposed.The dynamic priority of the task is updated in real time according to the initial priority of the task and the relative waiting time.(4)Aiming at problems such as workshop task assignment conflicts and slow demand response,the intelligent workshop and unmanned platform models were established in the Gazebo environment,and the multi-sensor fusion SLAM experiment,the path planning experiment based on spatiotemporal grid map and the task scheduling experiment based on dynamic priority were carried out respectively.Finally,the three are integrated to complete the intelligent workshop task scheduling experiment,which further verifies the effectiveness of the above method.