Robots are going to perform many household chores in the future, including wiping surfaces, vacuuming the floor, and sweeping dust. The performance of these robots is not always observable as dust and residual water may not be perceivable by computer vision. Daniel Leidner and Michael Beetz purpose using a haptic perception system to optimize wiping motion.
In this work we propose to utilize haptic perception paired with a qualitative effect representation to reason about the task performance of robotic wiping motions despite poor visual information. In particular, we relate the desired contact force to the measured end-effector force in order to simulate the effect of previously executed wiping motions. This way we are not just able to distinguish good from bad contact situations, but also replan recovery motions with regard to the effect-space to accomplish the commanded cleaning task.
As you can see in the video, these robots can be thought to recover from errors and make correct adjustments.
[HT]