Semiautonomous wheelchair human-aware navigation in crowded environments based on long-term pedestrians motion prediction

This paper addresses the problem of controlling an electric powered wheelchair in dynamic, crowded areas. Navigating the vehicle effectively in such environmental conditions requires two issues to be considered: Relatively long term prediction of pedestrians' motion and an influence of the vehicle motion on the people behaviour in its vicinity. The purpose of this study is to develop a method that allows the wheelchair navigating safely in dynamic environments by taking into account human-vehicle interactions. The method presented in this paper uses a deterministic, model-based prediction strategy to generate the wheelchair motion. The motion has the feature that is acceptable by the patient being transported on the wheelchair. Using the long-term pedestrians motion prediction the minimal collision risk control strategy is applied. The method performance was evaluated in a simulated environment. Relevant simulation scenarios are presented and discussed.