Research On Automatic Navigation Of Intelligent Agricultural Machinery Based On 4WIS-4WID Structure

With the continuous advancement of agricultural facilities and modernization,automatic navigation technology has gradually become one of the research hotspots of intelligent agricultural equipment.The agricultural machinery which working in narrow working roads in the facility agriculture environment should have higher path tracking accuracy to achieve the above goals.It is necessary to study high-precision automatic navigation control methods.This paper developed a four-wheel independent drive and four-wheel independent steering(4WID-4WIS)structure test prototype for facility greenhouses,and the path tracking control method was researched and improved.The specific work was as follows:Firstly,a test prototype and its control system meeting the requirements of the subject were developed.In order to improve the path tracking accuracy of intelligent agricultural machinery automatic navigation in the greenhouse,a dynamic look ahead pure pursuit model was proposed based on particle swarm optimization algorithm.The position and heading angle of the intelligent agricultural machinery in the greenhouse were obtained by using the Ultra Wide Band(UWB)module and the electronic gyroscope;besides obtain the lateral deviation and heading deviation of agricultural machinery through calculation.An evaluation function and quantitative analysis method was designed for lateral and heading deviation.The particle swarm optimization algorithm was used to select the optimal look ahead distance in the pure pursuit model in real time,and the parameters of the particle swarm optimization algorithm were improved in order to improve the efficiency of the algorithm.The effectiveness of dynamic look ahead pure pursuit model path tracking control was verified through MATLAB simulation tests.Secondly,a variable speed controller of the prototype was designed to improve the adaptability of path tracking control of speed.Speed detection was realized by fusing sensor information according to the steering motion model of the prototype,adaptive control was performed based on the lateral and heading deviation during the path tracking of the prototype.and a smooth speed planning method was proposed to eliminate the influence of excessive speed fluctuations on navigation control of the prototype.The effectiveness of variable speed control was verified through simulation tests.Finally,in order to test the actual control effect of the algorithm in this paper,a real-vehicle navigation test was carried out in a greenhouse environment,including the linear path and multi-segment path tracking experiments.When tracking a straight path in different initial states,the average deviation of path tracking using the algorithm in this paper was 24.4 cm,and the average stable distance was 318.1 cm.Combined with speed control,the average deviation was reduced by 18.6%,and the average stable distance was shortened by 6.1%.In the multi-segment path test,the average deviation of path tracking using in-situ steering to switch the route was 9.2 cm,while the average deviation of path tracking using traditional steering to switch the route was 20.9 cm.The test prototype in this paper has the characteristics of high steering flexibility and suitable for narrow facility environment.The test results proved that the control method proposed in this study can achieve high-precision path tracking and can meet the needs of agricultural machinery for automatic navigation operations in the greenhouse.It can provide technical reference for automatic navigation control of facility mobile robots...