Coevolution particle filter for mobile robot simultaneous localization and mapping
2005
This paper presents the implementation of particle filter (PF) combined with a coevolution mechanism derived from the competition model of ecological species for mobile robot simultaneous localization and mapping (SLAM). The new version of particle filters is termed coevolution particle filter (CEPF). In CEPF particles are clustered into species, each of which represents the posterior estimation of robot's pose or landmark locations and is superior to a single particle. Since the coevolution between the species ensures that the multiple distinct hypotheses can be estimated at the same time. And the number of particles can be adjusted adaptively over time according to the population growth model. In addition, by using the crossover and mutation operators in evolutionary computation, intra-species evolution can drive the particles move towards the regions where the desired posterior density is large. So a small number of particles can represent the desired density well enough to make precise posterior estimation. Experimental results show that CEPF is efficient for SLAM and indicate superior performance compared with those of the EKF and PF method.
Keywords:
- Mathematical optimization
- Particle
- Crossover
- Machine learning
- Small number
- Particle number
- Mathematics
- Simultaneous localization and mapping
- Artificial intelligence
- Monte Carlo localization
- Extended Kalman filter
- Particle filter
- Evolutionary computation
- Computer vision
- Estimation theory
- Mobile robot
- Algorithm
- Computer science
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
14
References
2
Citations
NaN
KQI