Dora, a Robot Exploiting Probabilistic Knowledge under Uncertain Sensing for Efficient Object Search

Dora, the robot, is trying to find object in its environment. Instead of just exhaustively searching everywhere, Dora is equipped with probabilistic reasoning, representations, and planning to exploit uncertain common-sense knowledge, such as that cornflakes are usually found in kitchens, while also accounting for the uncertainty of sensing in the real-world. Dora demonstrates how to combine task and observation planning in the presence of uncertainty by autonomously switching between contingent and sequential planning sessions. The demonstration emphasises the benefit of employing a robot with common-sense knowledge and the benefit of the switching planner.