Path planning and tracking control of orchard wheel mower based on BL-ACO and GO-SMC

• A mathematical model was constructed for the turning mode of lawn mowers in a low rootstock dense orchard. • Modify traditional ACO to make it suitable for solving this problem. • Based on SMC, GO-SMC was designed and its superiority was demonstrated. This research proposes an improved ant colony algorithm (BL-ACO) path planning algorithm and a tracking controller based on global optimal sliding mode variable structure control (GO-SMC) for the problem of path planning and tracking control of lawn mowers in quadrilateral orchard environments. The novelty of this research lies in two aspects. On one hand, we analyze the operating scenarios of lawn mowers in standardized orchards, then transform the path planning problem into a traveling salesman problem, and mathematically model the U-shaped and T-shaped turning strategies based on the characteristics of the wheeled lawn mower. In order to make the ant colony algorithm suitable for orchard operation path optimization problems, we modified its pheromone update rules, heuristic functions, state transition probabilities, and other equations. In order to accelerate the convergence speed of the ant colony algorithm, we use the bilayer ant colony algorithm optimization strategy. On the other hand, we establish a kinematic model with the wheeled lawn mower as the control object, and design a control law using a hyperbolic tangent function to ensure the global stability of the trajectory tracking control system. Furthermore, we demonstrate through Lyapunov stability analysis that the GO-SMC controller can ensure the mower tracks the reference path accurately. The simulation experiments of path planning and tracking control show that BL-ACO and GO-SMC perform the best compared to similar algorithms. Field experiments shows that BL-ACO & GO-SMC, with a time reduction rate of 47.58 % and a fuel consumption rate reduction of 47.59 % compared to line by line & SMC.