Comparison between Genetic Algorithms and Particle Swarm Optimization
EP '98 Proceedings of the 7th International Conference on Evolutionary Programming VII
Hi-index | 0.00 |
The position-velocity (PV) process model can be applied to the GPS Kalman filter adequately when navigating a vehicle with constant speed. However, when an abrupt acceleration motion occurs, the filtering solution becomes very poor or even diverges. To avoid the limitation of the Kalman filter, the particle swarm optimization can be incorporated into the filtering mechanism as dynamic model corrector. The PSO can be utilized as the noise-adaptive mechanism to tune the covariance matrix of process noise and overcome the deficiency of Kalman filter. In this paper, PSO-aided Kalman filter approach is employed for tuning the covariance of the GPS Kalman filter so as to reduce the estimation error during substantial maneuvering. Performance evaluation for the PSO-aided Kalman filter as compared to the conventional Kalman filter is provided.