Readings in qualitative reasoning about physical systems
Readings in qualitative reasoning about physical systems
Gross motion planning—a survey
ACM Computing Surveys (CSUR)
Computational research on interaction and agency
Artificial Intelligence - Special volume on computational research on interaction and agency, part 1
Behaviorist intelligence and the scaling problem
Artificial Intelligence
Hi-index | 0.00 |
This article describes a novel qualitative navigation method for autonomous wheelchair robots in typical home environments. The method accepts as input a line diagram of the robot environment and converts it into an enhanced grid in which qualitative representations of variations in sensor behavior between adjacent regions in space are stored. An off-line planner uses these representations to store at each grid cell appropriate motion commands that will ideally move the wheelchair in and out of each room in a typical home environment. An online controller accepts as input this enhanced grid along with a starting and goal position for the robot. It then compares the actual behavior of the sensors with the one stored in the grid. The results of this comparison are used to estimate the current position of the robot, to retrieve the planner instructions and to combine these instructrions with appropriate risk avoidance behaviors during navigation. This method has been tested both in simulation and as one of the subsystems on a prototype for an autonomous wheelchair robot. Results from both trials are provided.