Title :
Variational Bayesian data fusion of multi-class discrete observations with applications to cooperative human-robot estimation
Author :
Ahmed, Nisar ; Campbell, Mark
Author_Institution :
Sibley Sch. of Mech. & Aerosp. Eng., Cornell Univ., Ithaca, NY, USA
Abstract :
A new method is presented for fusing conventional continuous sensor observations with discrete multi-categorical state-dependent information, which can be furnished by humans in many cooperative human-robot interaction problems. The hybrid likelihood function for mapping between continuous hidden states and categorical observations are specified via softmax models. Although softmax models avoid discretization of continuous states, they are challenging to implement for real-time data fusion since they are not analytically integrable. An approximation based on variational Bayesian (VB) methods is presented here to obtain fast closed-form Gaussian solutions to the desired posteriors in cases where the hidden continuous states have Gaussian pdfs. A joint human-robot target localization example illustrates the properties and utility of the VB hybrid fusion strategy, which also applies more generally to inference in hybrid Bayesian networks and mixture models.
Keywords :
Bayes methods; Gaussian processes; human-robot interaction; sensor fusion; variational techniques; closed-form Gaussian solution; continuous sensor observations; cooperative human-robot estimation; cooperative human-robot interaction problem; discrete multicategorical state-dependent information; hybrid Bayesian network; hybrid likelihood function; joint human-robot target localization; mixture model; multiclass discrete observations; softmax model; variational Bayesian data fusion; Bayesian methods; Humanoid robots; Humans; Orbital robotics; Position measurement; Robot kinematics; Robot sensing systems; Robotics and automation; State estimation; USA Councils;
Conference_Titel :
Robotics and Automation (ICRA), 2010 IEEE International Conference on
Conference_Location :
Anchorage, AK
Print_ISBN :
978-1-4244-5038-1
Electronic_ISBN :
1050-4729
DOI :
10.1109/ROBOT.2010.5509521