Real-time implementation of Bayesian models for multimodal perception using CUDA

In this text we present the real-time imple- mentation of a Bayesian framework for robotic multisen- sory perception on a graphics processing unit (GPU) using the Compute Unified Device Architecture (CUDA). As an additional objective, we intend to show the benefits of parallel computing for similar problems (i.e. probabilistic grid-based frameworks), and the user-friendly nature of CUDA as a programming tool. Inspired by the study of biological systems, several Bayesian inference algorithms for artificial perception have been proposed. Their high computational cost has been a prohibitory factor for real- time implementations. However in some cases the bottle- neck is in the large data structures involved, rather than the Bayesian inference per se. We will demonstrate that the SIMD (single-instruction, multiple-data) features of GPUs provide a means for taking a complicated framework of relatively simple and highly parallelisable algorithms operating on large data structures, which might take up to several minutes of execution with a regular CPU imple- mentation, and arrive at an implementation that executes in the order of tenths of a second. The implemented multi- modal perception module (including stereovision, binaural sensing and inertial sensing) builds an egocentric repre- sentation of occupancy and local motion, the Bayesian Volumetric Map (BVM), based on which gaze shift deci- sions are made to perform active exploration and reduce the entropy of the BVM. Experimental results show that the real-time implementation successfully drives the robotic system to explore areas of the environment mapped with high uncertainty.