Towards Object Mapping in Non-Stationary Environments with Mobile Robots

We propose an occupancy grid mapping algorithm for mobile robots operating in environments where objects change their locations over time. Virtually all existing environment mapping algorithms rely on a static world assumption, rendering them inapplicable to environments where things (chairs, desks,...) move. A natural goal of robotics research, thus, is to learn models of nonstationary objects, and determine where they are at any point in time. This paper proposes an extension to the well-known occupancy grid mapping technique. Our approach uses a straightforward map differencing technique to detect changes in an environment over time. It employs the expectation maximization algorithm to learn models of non-stationary objects, and to determine the location of such objects in individual occupancy grid maps built at different points in time. By combining data from multiple maps when learning object models, the resulting models have higher fidelity than could be obtained from any single map. A Bayesian complexity measure is applied to determine the number of different objects in the model, making it possible to apply the approach to situations where not all objects are present at all times in the map.