Search in the Context of Complex Robot-World Interaction

In contrast to typical planning problems, automated earthmoving operations such as digging a trench or leveling a mound of soil pose some additional challenges. First, soil is diffuse and therefore a unique description of the world requires a very large number of variables. Second, the interaction between the robot and the world is very complex to the point that it is not possible to accurately predict the effect of a candidate action, in this paper, 1 pose planning for earthmoving operations as a problem of constraint optimization. I discuss why on-line search is necessary and present experimental results from a testbed where a robot is tasked to dig a trench.