Don't do things you can't undo: reversibility models for generating safe behaviours
Kruusmaa, M.; Gavshin, Y.; Eppendahl, A. (2007). Don't do things you can't undo: reversibility models for generating safe behaviours. 2007 IEEE International Conference on Robotics and Automation; Roma; 10-14 April, 2007. Roma: IEEE, 1134−1139.
Kruusmaa, M.; Gavshin, Y.; Eppendahl, A.
2007 IEEE International Conference on Robotics and Automation; Roma; 10-14 April, 2007
3.1. Artiklid/peatükid lisas loetletud kirjastuste välja antud kogumikes (kaasa arvatud Thomson Reuters Book Citation Index, Thomson Reuters Conference Proceedings Citation Index, Scopus refereeritud kogumikud)
M1 - Copyright 2007, The Institution of Engineering and Technology; U1 - 9561695; U2 - reversibility models; U2 - safe behaviour generation; U2 - robots; U2 - obstacle avoidance, We argue that an ability to determine the reversibility of actions allows a robot to identify safe behaviors autonomously. We introduce a notion of reversibility model and give a definition of model refinement. We implement this on a real robot and observe that, when a reversibility model is refined by the addition of proximity sensors, obstacle avoidance emerges as a side-effect of avoiding irreversible actions. We interpret this as evidence of a deep connection between reversibility and safe behaviour. We also observe that, on the real robot, reversibilities are learned as efficiently as a dedicated reward function. We conclude that reversibility identification may provide an abstract and yet practical method of generating a variety of safe behaviours.
intelligent robots, learning (artificial intelligence), self-adjusting systems