Multiple Population Genetic Algorithm-Based Inverse Kinematics Solution for a 6-DOF Manipulator

Compared to traditional fixed configuration manipulators, modular manipulators occupy less space, offer greater flexibility, and demonstrate stronger adaptability to diverse environments. These characteristics make them particularly suitable for operating in unknown environments, such as disaster rescue and pipeline inspection. This paper presents the design of a modular robotic arm and proposes a novel approach to solving the inverse kinematics problem for a 6-DOF (degree of freedom) tandem manipulator using a Multi-population Genetic Algorithm (MPGA). The proposed method overcomes the high nonlinearity and computational complexity of traditional genetic algorithms (SGA) by incorporating real-number encoding, Exponential Ranking Selection, and a combination of Simple and Gaussian mutations. These improvements significantly enhance the algorithm's convergence speed, accuracy, and robustness, making it suitable for complex robotic systems. The manipulator's forward kinematics is established using the Denavit-Hartenberg (D-H) method, and the MPGA optimizes the inverse kinematics solution. Simulations and experiments on both fixed and mobile platforms demonstrate the MPGA's superior performance in terms of computational efficiency and solution accuracy. The manipulator accurately followed the planned trajectory, validating the method's effectiveness. This study provides a novel and efficient solution for inverse kinematics in high-DOF manipulators, offering potential applications across various robotic systems.