Toward a Generalized Risk Assessment Method on Occupancy Grids

To navigate safely and efficiently, a robot must be aware of its surroundings. Although Bayesian occupancy grids are an efficient way to map the environment, they are not wellsuited to assess risks of a specific path. In this paper, we extend the work of Laconte et al. [1] to generate meaningful risks in 3D environments. Here, we define the risk as the deformation energy of the robot's wheels due to collisions.