Cognitive Force Control of Excavators

Abstract

One of the high‐volume and repetitive construction operations on earth is the excavation of soil. Automation of planetary excavation work requires a system that is able to perform the planned digging work that is responsive to interaction forces experienced during excavation. The conditions and importance of excavation motivates the goal of having a robot excavator that would be given the location and dimensions of a desired excavation and would plan the excavation process, sense the environment, react to changes as excavation proceeds, and accomplish the desired excavation with almost no external or human intervention. Cognitive robots require information for modeling, planning, and acting at different levels. The development of such a control system requires a dynamic model to describe the evolution of the excavator motion with time. In planetary digging, the regolith resists bucket movements, demanding the excavator to be able to comply with the restrictions imposed by the environment. A control system that is able to perform a preplanned digging work while complying with the regolith restrictions is described in the present paper.