Rehabilitation Engineering and Electromechanical Design Lab | |||||||
|
Human Force InteractionsProject Background
This research seeks to mesh humans and robots more effectively to improve our quality of life. Our contribution toward this broad goal is investigating how we sense and interact with forces, other humans, and robots. A motivating example from our work is a project where we are quantifying treatment given by physical therapists in the medical field so patients get more consistent and accurate treatment to speed up recovery. This is part of a much broader, long standing effort to provide evidence based healthcare. In order to move toward that goal of comparing treatments based on qualitative data, we must first understand the current state of treatment. This project is about doing just that in physical therapy using robotics. Another area of focus for us is understanding how humans control robotic systems. So, let's say that in 14 years that you want to have a firefighter piloting a swarm of fire extinguishing drones: taking the best of both the human's ability to contextualize new information and react and the robot's indefatigable, fearless, and precise characteristics. To do that, you need virtual interfaces and control systems that work with the pilot. Our research is about understanding the nuances of those types of interactions to enable the next wave of robotic technology like fire fighting drone swarms. Technical Details
The first piece investigates how humans can control position through self-generated forces during dynamic tasks. We have explored several factors including how the speed of the task, the direction of perturbing forces, and the perturbation profile affect the position accuracy in a task. As mentioned above, we are evaluating the dynamic motor skills of physical therapists using a task that approximates patient-assistive interactions, which are essentially position control motions using self-generated forces. Another piece inspects how humans control complicated systems. A series of cart inverted pendulums are used to investigate this relationship where the task is to balance the inverted pendulum by controlling the cart. We are investigating how performance is affected by a number of consequential factors such as the system's degrees of freedom, the control interface, and the human-computer or human-machine interface. Finally, on a more fundamental level, we are studying several important factors affecting the memory of force perceptions in discrimination tasks. The effects of the inter-stimulus interval and the persistence of a force perception memory are individually examined. We are applying this research to our other research in physical therapy. Related Publications
This work was supported by the National Science Foundation under Grant IIS-1319802. |
Last modified on Apr 30, 2020. |