Design And Implementation Of Multi-sensor Unmanned Ship Autonomous Navigation System Based On Path Planning

Based on ships,unmanned ships carry various special functional equipment,including control equipment and communication equipment,and can perform special operations in various waters.Due to the characteristics of high flexibility,small size,and wide range of uses,unmanned ships play a huge role in the military and civilian fields,and countries all over the world are competing for research.Based on this background,this subject designs and implements a multi-sensor unmanned ship autonomous navigation control system based on path planning,which realizes the communication between the various parts of the unmanned ship and the autonomous navigation of the unmanned ship.First of all,on the basis of reading a large number of domestic and foreign documents,the research status and future development trends of unmanned ships and path planning are introduced,and the basic research route of this subject is proposed.The overall design plan of the unmanned ship is designed,and the entire autonomous navigation control system of the unmanned ship is divided into three parts,namely the remote monitoring system,the Raspberry Pi master control system and the STM32 bottom control system.Aiming at the autonomous path planning of unmanned ships,an overall path planning scheme based on the ROS navigation framework is designed.Secondly,the positioning of the unmanned ship is the key to path planning.This article uses GPS/IMU integrated navigation to locate the unmanned ship,and uses the Kalman filter algorithm to fuse and output the GPS/IMU data,which can overcome GPS alone.Insufficient positioning.After the positioning is completed,the third-party map and Gmapping algorithm are used to construct the unmanned ship water area map,and the nearest neighbor clustering algorithm is used to process the raw data collected by the lidar to obtain more accurate obstacle location information.On the basis of precise positioning and high-precision water map,this project adopts the unmanned ship hybrid path planning scheme that combines A* and DWA algorithms,and conducts simulation analysis on A* and DWA algorithms.Then,according to the overall design of the above-mentioned unmanned ship’s systems and the design of positioning mapping algorithm and path planning algorithm,the software and hardware platform of the unmanned ship’s autonomous navigation control system is built.Starting from the application of inland river scenes,the hull structure and mathematical model of the unmanned ship are designed and deduced.The remote monitoring system runs on the PC for remote real-time monitoring of the status of the unmanned ship and the issuance of target points;the Raspberry Pi main control system is designed based on the ROS system and mainly runs the driver programs and positioning and navigation algorithms of each node;the bottom layer of STM32 The control system is designed based on Free RTOS,responsible for the motor PID control,and realizes the collection and upward feedback of various sensor data.Finally,use the unmanned ship’s physical platform built in this topic to conduct autonomous navigation experiments on the water surface.From the experimental results,it can be seen that the unmanned ship can navigate along the planned route more accurately.During the navigation,although there is a sharp turn,it tends to be stable on the whole,which verifies the unmanned ship ’ s software and hardware platform.Function and feasibility of path planning algorithm.The platform is compact and flexible,with complete basic functions,which provides a useful reference for the design of inland unmanned ships.