Position Estimation on Image-Based Heat Map Input using Particle Filters in Cartesian Space

This paper presents an approach for using an image-based heat map as measurement input of a particle filter. Pixels of the heat map are transformed into Cartesian space relative to the robot and regarded as single measurements. The approach uses a novel observation model to weight the particles accordingly to the heat map pixels. While this paper focuses on handling FCNN output, the method is also applicable to other feature recognition methods like saliency approaches. The proposed method shows similar performance in standard cases and huge improvements on erroneous input compared to the conventional approach.