CRF-Filters: Discriminative Particle Filters for Sequential State Estimation

@inproceedings{citeulike:2822219, abstract = {Particle filters have been applied with great success to various state estimation problems in robotics. However, particle filters often require extensive parameter tweaking in order to work well in practice. This is based on two observations. First, particle filters typically rely on independence assumptions such as "the beams in a laser scan are independent given the robot's location in a map". Second, even when the noise parameters of the dynamical system are perfectly known, the sample-based approximation can result in poor filter performance. In this paper we introduce CRF-filters, a novel variant of particle filtering for sequential state estimation. CRF-filters are based on conditional random fields, which are discriminative models that can handle arbitrary dependencies between observations. We show how to learn the parameters of CRF-filters based on labeled training data. Experiments using a robot equipped with a laser range-finder demonstrate that our technique is able to learn parameters of the robot's motion and sensor models that result in good localization performance, without the need of additional parameter tweaking.}, author = {Limketkai, B. and Fox, D. and Liao, Lin }, booktitle = {Robotics and Automation, 2007 IEEE International Conference on}, citeulike-article-id = {2822219}, doi = {http://dx.doi.org/10.1109/ROBOT.2007.363957}, journal = {Robotics and Automation, 2007 IEEE International Conference on}, keywords = {machine-learning}, pages = {3142--3147}, posted-at = {2008-05-22 06:19:38}, priority = {2}, title = {CRF-Filters: Discriminative Particle Filters for Sequential State Estimation}, url = {http://dx.doi.org/10.1109/ROBOT.2007.363957}, year = {2007} }

TY - CONF ID - citeulike:2822219 L3 - citeulike-article-id:2822219 N2 - Particle filters have been applied with great success to various state estimation problems in robotics. However, particle filters often require extensive parameter tweaking in order to work well in practice. This is based on two observations. First, particle filters typically rely on independence assumptions such as "the beams in a laser scan are independent given the robot's location in a map". Second, even when the noise parameters of the dynamical system are perfectly known, the sample-based approximation can result in poor filter performance. In this paper we introduce CRF-filters, a novel variant of particle filtering for sequential state estimation. CRF-filters are based on conditional random fields, which are discriminative models that can handle arbitrary dependencies between observations. We show how to learn the parameters of CRF-filters based on labeled training data. Experiments using a robot equipped with a laser range-finder demonstrate that our technique is able to learn parameters of the robot's motion and sensor models that result in good localization performance, without the need of additional parameter tweaking. JF - Robotics and Automation, 2007 IEEE International Conference on EP - 3147 TI - CRF-Filters: Discriminative Particle Filters for Sequential State Estimation T2 - Robotics and Automation, 2007 IEEE International Conference on SP - 3142 KW - machine-learning AU - Limketkai, B AU - Fox, D AU - Liao, Lin PY - 2007/// UR - http://dx.doi.org/10.1109/ROBOT.2007.363957 ER -