Improved Simultaneous Localization and Mapping Estimation using Crow Search Algorithm Based Particle Filter

Trajectory tracking and positioning are essential requirements in many areas, including robots and autonomous vehicles. In some cases, such as in areas where GPS signals are weak or not available, trajectory tracking is used as an alternative positioning system. In these cases, simultaneous localization and mapping (SLAM), is of great importance as it does not require prior knowledge and empirical offline fingerprint. SLAM can be combined with signal processing algorithms among which, particle filter stands out. However, challenges exist such as particle weights degradation and particles impoverishment that need to be dealt with. In fact, the loss of particle diversity for estimation has led to the lack of particles. To overcome this problem, one solution is to diversify the selection of particles after resampling. In this paper, we proposed a crow search algorithm (CSA) to overcome these issues and improve position estimation. The simulation results showed that this algorithm greatly improved the performance of fast SLAM.