Robotic arm time-optimal trajectory planning using the Adaptive Gold Search algorithm

To address issues of local optimization, slow convergence, and poor accuracy, this work introduces a time-optimal trajectory planning method using the Adaptive Gold Search algorithm (AGS). The method enhances algorithm diversity and escapes local optima through elite reverse strategy population initialization. It employs an adaptive nonlinear decreasing factor for a balance between linearity and nonlinearity, improving accuracy and convergence speed. The golden sinusoidal variation technique enhances global search capabilities and robustness in robot arm trajectory planning. Experimental results show that the AGS algorithm produces smooth joint angular displacement, acceleration, velocity, and jerk curves. It outperforms the GA, PSO, GOOSE, and GWO algorithms in time optimization by 13.31%, 13.61%, 5.34%, and 4.73%, respectively, effectively reducing trajectory planning time.