Study On Path Planning Of Orchard Mowing Robot Based On GPS And Inertial Navigation

With the rapid development of intelligent machines,mobile robots have been widely used in the field of agriculture.The level of intelligent mechanization for orchard area management is also constantly improving,but the work of weeding in the garden mainly depends on hand-held and riding lawn mowers.In order to improve the level of intelligent mechanization of the weeding work in the orchard area,effectively improve the efficiency of the work,and free people from the heavy labor,this paper designs and studies the orchard mowing robot based on GPS and inertial navigation system,and designs a complete coverage path planning strategy And trajectory tracking algorithm to realize simulation experiment of algorithm in simulated orchard environment.1)Research on mowing robot control system and positioning.Design the control system and positioning system of the mowing robot,and build a complete hardware system;Implement GPS differential technology by CORS mode to receive the network correction signal.For the unstable positioning effect when GPS and IMU work alone,used Kalman filter performs multi-sensor fusion filtering of GPS,IMU and encoder;the positioning function of the mowing robot is used to identify the orchard boundary and the position of the orchard to realize the establishment of an orchard electronic map model.2)Research on full coverage path planning algorithm.The main path planning methods are studied,combined with the orchard area environmental constraints,the orchard application problem of the reciprocating algorithm is analyzed,and the orchard area full coverage path planning strategy is proposed;there are two situations for full area coverage,using sub-region and mechanical obstacle avoidance The reciprocating path planning method realizes the full area coverage traversal of the orchard area.3)Research on trajectory tracking algorithm.Under the premise that the target path is known,based on the trajectory characteristics of the full coverage path,the problems of the pure tracking algorithm are analyzed,and the number of parameters is used to select the preview point for multiple preview points in the trajectory tracking process;for the lateral error When the range is large,a strategy of introducing geometric area parameters is proposed to select the preview point.When the lateral deviation is less than a certain value,the foresight distance is converted into the original foresight distance,the speed of trajectory convergence is increased,and simulation verification is performed.4)Experimental research for simulating orchard environment.The complete coverage path planning algorithm proposed in Chapter 3 is used for path planning of the orchard electronic map,and two orchard environments are designed according to the spacing of the fruit trees.Compared with the traditional reciprocating path planning method,the first case is the regionalized reciprocating method The missing cut rate is reduced by 2.1%,the repetition rate is reduced by 1.7%,the second case of the regionalized reciprocating method,the missing cut rate is reduced by 2.5%,and the repetition rate is reduced by 2.47%;then the proposed improved pure tracking algorithm is fully covered The tracking simulation is compared with the traditional pure tracking algorithm.The results show that compared with the traditional pure tracking algorithm,the improved pure tracking algorithm has a wider angular range of 0.2rad and a lateral error of 0.063 m,which effectively reduces the tracking error.