Research On Obstacle Avoidance System Of Intelligent Weeder In Camellia Oleifera Forest

In view of the complex weeding operation environment in the hilly and mountainous areas of the south,in order to further reduce the labor intensity of the staff,the intelligent upgrading of agricultural equipment is realized.This paper further studies based on the original research of the research group,taking the intelligent lawn mower independently developed by the project team as the research object,and using binocular camera,ultrasonic sensor and STM32 control panel and other intelligent control hardware to carry out the following research.(1)Based on the forest weeder independently developed by the project team,the obstacle avoidance system adapted to the forest land is designed,the left and right motor drive mode is used to control the overall steering of the woodland intelligent lawn mower,the skateboard structure is used to assist obstacle avoidance,the electromagnetic hydraulic valve is used to control the weeding of the lawnwer,and the selected ultrasonic sensor,binocular camera,control chip and its peripheral module are selected.In order to solve the planning of obstacle detour and obstacle avoidance path within the local feasible range of forest weeder.Firstly,based on the results of the optimized fuzzy adaptive adaptive rule table and the improved gravitational function model and repulsive function model of the artificial potential field method,a reasonable obstacle avoidance planning path is carried out for the forest weeder.(2)Firstly,the fuzzy adaptive rule is used to calculate and plan the safety distance of woodland weeders and camellia oleifera trees,and then the improved gray wolf algorithm is used to optimize the fuzzy adaptive rules to solve the optimal obstacle avoidance strategy during the movement of woodland weeders.Then,based on the results of the fuzzy adaptive rule table optimized by the gray wolf algorithm,the improved artificial potential field method algorithm based on the optimized fuzzy adaptive rule,the improved gravitational function model and repulsive function model of the artificial potential field method are used to carry out a reasonable obstacle avoidance planning path for the forest weeder,and the simulation results of the improved artificial potential field method are analyzed.The time for the woodland intelligent weeder to make decisions when identifying obstacles while running is reduced to 2.3416 seconds,which provides a guarantee for the safety of the woodland intelligent weeder and Camellia oleifera trees.(3)Based on the ant colony algorithm,the search mode of A*algorithm is used to simulate the path planning of woodland intelligent weeder in uneven hilly and mountainous areas,and the heuristic function,transfer probability and angle factors of ant colony algorithm are optimized.The simulation results show that the improved ant colony algorithm is adopted,the search speed is fast,the optimal path is found after the number of iterations reaches 32 times,and the operation is completed in 2.8964 seconds,which meets the needs of planning the path while the forest weeder is running.(4)In order to verify the effect of the above algorithm and obstacle avoidance strategy,it was tested in the experimental forest land of Camellia oleifera fruit for many times,and the terrain,obstacles,obstacle avoidance algorithms and strategies were further studied.This paper verifies the obstacle avoidance system,obstacle avoidance strategy and obstacle avoidance path planning of woodland intelligent weeden mower designed in this paper,which can well meet the safety and planning of woodland intelligent weeder in hilly and mountainous areas.