Active perception strategy for vehicle localisation and guidance

Vehicle localisation in outdoor environment is an important issue. When this localisation system has to provide an accurate and reliable position for an automatic guidance system, this is a challenge. In this paper, we propose a supervised active localisation system to satisfy this need. When localisation system mustn't meet vehicle control process requirements: accurate position and high confidence level, the supervisor requests the active detection of a particular landmark. It must be seen as a first step towards the development of a multi-sensor localisation system. The main contributions of this paper are: 1) The introduction of the notion of the perceptive triplet within a general framework that associates landmarks, sensors and detectors to supervise the detections. 2) This is the supervisor that determines each time which landmark, with which sensor and detector should be used to detect this landmark in order to best improve the localisation. The supervisor constitutes the intelligent part of this localisation system. It decides when it's necessary to detect a landmark. Our system was tested in an outdoor environment, where it succeeded in accurately localising the vehicle during automatic guidance.