An improved particle swarm optimization algorithm and its application in solving forward kinematics of a 3-DoF parallel manipulator

Abstract

This paper proposes an improved particle swarm optimization to study the forward kinematic of a solar tracking device which has two rotational and one translational degree of freedom. The forward kinematics of the parallel manipulator is transformed into an optimization problem by solving the inverse kinematics equations. The proposed method combines inertial weight with the iterations number and the distance between current swarm particles and the optimum to improve convergence ability and speed. The novel cognitive and social parameters are adjusted by the inertia weight to enhance unity and intelligence of the algorithm. A stochastic mutation is used to diversify swarm for faster convergence via local optima evasion in high dimensional complex optimization problems. The performance of the proposed method is demonstrated by applying it to four benchmark functions and comparing convergence with three popular particle swarm optimization methods to verify the feasibility of the improved method. The behaviors of the proposed method using variable cognitive and social parameters and fixed value are also tested to verify fast convergence speed of variable parameters method. And further, an application example uses the method to determine the forward kinematics of a three-degree-of-freedom parallel manipulator. Finally, the mechanism simulations model of the parallel manipulator are carefully built and analyzed to verify the correctness of the proposed algorithm in PTC Creo Parametric software. In all cases tested, the proposed algorithm achieved much faster convergence and either improved or proximal fitness values.