Towards Mapping Dynamic Environments

We propose an algorithm for mapping dynamic environments. Our algorithm creates a representation of the environment where objects move over time. Our approach is based on maintaining two occupancy grids in parallel. One grid models the static parts of the environment, and the other models the dynamic parts of the environment. The union of the two provides a complete description of the environment over time. We also apply a size-based classifier to the cells in the dynamic map, to detect and track moving objects. Results in simulation and with physical robots show a reasonable rate of detection of moving objects.