Neural network PID steering control method for hydrostatic drive tracked vehicle

Abstract

The invention discloses a neural network PID steering control method for a hydrostatic drive tracked vehicle, which comprisesthe following steps: S1, reading a driver target relative steering radius rho ref and a target vehicle speed vref adjusted by a steering control coordination control strategy, S2, determining whether a driver vehicle is driving straightly, if yes, executing a straight driving control subprogram, and inputting the result into a neural network PID controller to be calculated, adjusting and sending pump and motor displacement instructions, and if the result is not the straight driving, judging R to be equal to 0.5 B. According to the control method, after self-adaptive adjustment of the neural network is added, the problem that adjustment of a speed adjusting scheme lags behind is solved, the steering controller can pre-judge the steering state of the vehicle, and the target vehicle speed of a driver is corrected in advance; the corrected input signal of the driver can enable the vehicle to quickly and safely steer under the conditions of no sideslip, no slipping and system pressure meeting the maximum pressure limit.

Description

technical field [0001] The invention relates to a vehicle steering control method, in particular to a neural network PID steering control method of a hydrostatically driven crawler vehicle, and belongs to the technical field of vehicle steering. Background technique [0002] Tracked vehicles have the advantages of low ground pressure, good off-road performance, and strong ability to climb and overcome obstacles. Flexible upper layout is the main development direction of tracked vehicle transmission in the future. Steering ability, as an important indicator of combat skills, is especially important for tracked vehicles driven by hydrostatic fluid. The steering of the hydrostatic drive tracked vehicle is accomplished by controlling the displacement of variable pumps and variable motors on both sides, so that the driving wheels on both sides of the steering mechanism output different speeds, thereby driving the tracks on both sides to produce a speed difference. The steering ...

AI Technical Summary

Problems solved by technology

[0004] The object of the present invention is to provide a neural network PID steering control method for hydrostatically driven crawler vehicles, to solve the limitations of the hydraulic motor output torque and engine output power proposed in the above-mentioned background technology, which will inevitably lead to slow response speed of steering and irregular steering process. Stability, poor trajectory controllability issues

Images