An improved particle swarm optimization(PSO)algorithm is proposed to address the problems that conventional PSO algorithms are prone to local optima

late-stage search stagnation

and low convergence efficiency in solving robot trajectory planning.Individual evolution speed and population evolution speed parameters are introduced

and an adaptive inertia weight function is proposed by combining these two parameters with the current number of iterations.In addition

the update strategy of the improved algorithm is modified.The mutation mechanism of genetic algorithm is integrated to endow particles in poor search areas with a certain probability of sudden jumps.Finally

simulation experiments are conducted

and the improved algorithm is applied to a humanoid touch robotic arm.The results show that the improved PSO algorithm can ensure that the robot meets kinematic constraints during motion

with the speed and acceleration remaining stable without sudden changes.It converges faster than conventional PSO algorithms

and the interpolation time is shortened by nearly 30%

effectively improving the operation efficiency of the robot.