Inverse kinematics (IK) is an optimization problem solving the path or trajectory a multi-jointed body should take for an extremity to reach a specified target location. When also considering the flow of movement, IK becomes a multi-objective optimization problem (MOP). This study proposes the use of the vector evaluated particle swarm optimization (VEPSO) algorithm to solve IK. A 3D character arm, with 7 degrees of freedom, is used during experimentation. VEPSO's results are compared to single-objective optimizers, as well as an optimizer that uses weighted aggregation to solve MOPs. Results show that the weighted aggregation approach can outperform IK-VEPSO if the correct weight combination (that is problem dependent) has been selected. However, IK-VEPSO produces a set of possible solutions.
展开▼
