Obstacle avoidance path planning for autonomous navigation agricultural machinery

Objective  To realize static obstacle avoidance of autonomous navigation agricultural machinery when it operates in the field.

Method  Two path planning algorithms of obstacle avoidance were proposed under known working environment. The single obstacle avoidance algorithm was proposed based on the movement rule of agricultural machinery and according to the position and the size of the obstacle. On the basis of the single obstacle avoidance algorithm, according to the size of safe driving area, the double/multiple obstacles avoidance algorithm was proposed according to left and right obstacle avoiding strategies.

Result  When the single obstacle was located in different locations and the speed of agricultural machinery was 0.3 m·s⁻¹, compared with L algorithm, driving path reduced by 35%, 26%; Accumulative error of driving path reduced by 53%, 82%; Variance reduced by 64%, 66%. When the driving speed was 0.5 m·s⁻¹, driving path reduced by 38%, 22%; Accumulative error reduced by 66%, 26%; Variance reduced by 41%, 71%. When the speed of agricultural machinery was 0.3 and 0.5 m·s⁻¹ under the condition of multiple obstacles, accumulative tracking errors of driving path was 9.99, 4.13 m, and variances were 0.022 1, 0.027 0 m², respectively.

Conclusion  The proposed algorithm has some advantages in driving path, accumulative error of driving path, stability of theoretical path tracking and adaptability to road condition.