Conjugate Unscented Particle Filter Based Monte Carlo Localization for Mobile Robots

Jiang Xue Fei,Song Yu
DOI: https://doi.org/10.4028/www.scientific.net/amm.556-562.2266
2014-01-01
Applied Mechanics and Materials
Abstract:For mobile robot localization in known environment, the 5th-order Conjugate Unscented Particle Filter Monte Carlo Localization (CUPF-MCL) algorithm is proposed. CUPF-MCL combines the 5th-order Conjugate Unscented Transform (5th CUT) with Kalman Filter to generate more accuracy particle filter proposal distribution, calculating the transition density up to the 5th-order nonlinearity. In simulation, the performance of CUPF-MCL is compared with that of dead reckoning, PF-MCL, EPF-MCL and UPF-MCL. Results show that CUPF-MCL improves the accuracy of localization.
What problem does this paper attempt to address?