The auction technique for the sensor based navigation planning of an autonomous mobile robot (Q1383989)

An algorithm for motion planning on a completely unknown work space is presented. The main features of the method are: 1. an iterative solution of the shortest path problem which alternates with the acquisition of new knowledge by a sensor system; 2. the use of an auction technique to solve the optimization problem quickly; 3. the updating of the weights in a scheme similar to artificial potential methods. The proposed method combines the advantages of both graph optimization and the artificial potential method, avoiding the high complexity of the former and the possibility of local minima of the latter.