Paper
XPROBE: An Experimental System for Programming Robots by Example
Abstract
XPROBE extends the teaching-by-showing method of robot programming to the teaching of sensor strategies. The XPROBE system has a dialogue with the user, during which the user is directed to guide the robot through an example of the desired task. XPROBE then automatically writes an AML program that includes sensory decision making as a generalization of the taught example. © 1984, Sage Publications. All rights reserved.
Related
Conference paper
The interplay of emotions and norms in multiagent systems
Conference paper
Hybrid reinforcement learning with expert state sequences
Conference paper