The Design And Implementation Of The Wheel-track Compound Unmanned Mobile Platform Control System

In this dissertation,a wheel-tracked composite unmanned mobile platform is taken as the research object,and the hardware and software design of the control system is developed.The overall framework of the control system is accomplished by distributed deployment of the superior and inferior systems.A multi-sensor information fusion localization algorithm based on improved extended Kalman filter is proposed.Real-time driving navigation is realized by combining the open interface of Baidu map route planning.The CamShift visual tracking algorithm is improved with the Kalman filter and deployed in the vehicle two-axis PTZ module to realize auxiliary sighting.Finally,the Qt Creator software is used to design the terminal integrated host computer to realize integrated integration of system control functions,status monitoring and user interaction.This dissertation firstly investigates the development trend and research status of unmanned ground combat vehicles at home and abroad,then introduces some classic models and series of combat robots in detail,and analyzes the key technologies and difficulties of mobile robots.On the basis of the existing research,the structural design scheme of the existing wheel-tracked composite unmanned mobile platform is introduced,and the kinematics model is established for the wheel drive and the track drive respectively.Completing the hardware and software framework design of the control system with superior and inferior systems distributed deployment.The inferior system machine works on STM32F4 single-chip microcomputer to realize the basic motion control of the mobile platform and the two-axis PTZ module and the information transmission with the on-board host computer.The on-board host computer adopts NVIDIA as the core controller.Jetson TX2 runs the robot operating system(ROS)in the Ubuntu system environment to realize core functions such as sensor information collection and processing,positioning algorithm deployment,visual auxiliary tracking and tracking algorithm deployment,and two-way communication with lower system computer.Remote server is responsible for remote control interaction.For the mobile robot in the pose tracking process,there is a problem that the single sensor or multi-sensor measurement system has limited ability to process environmental information.Combined with the extended Kalman filter algorithm,the sensor measurement information is analyzed and integrated,and the observation matrix is extended.Local update of the sensor to the state vector.Based on the established sensor and robot motion model,the experimental verification is carried out through the composite mobile platform.Theoretical analysis and experimental results show that the proposed method improves the generalization ability of the algorithm for different sensor types and sensors while ensuring the accuracy of positioning,and enhance the accuracy and flexibility of the measurement system.In the process of visual recognition and tracking of target objects using CamShift algorithm,when the target is occluded,the target moves too fast or the same color objects meet,the search window is susceptible to interference,and the tracking process is unstable or even interrupted.In this dissertation,the Kalman filter is used to improve the algorithm.The two-axis PTZ is used to realize the auxiliary sight-seeing function,and the feasibility of the algorithm is verified by experiments.Using Qt software to design the terminal integrated host computer to realize integrated integration of system control functions,status monitoring and user interaction.