For robots to be truly flexible, they need to be able to learn to adapt to partiallyknown or dynamic environments, to teach themselves new tasks, and to compensate for sensor and effector defects. The problem of robot learning has been an intensively studied research topic over the last decade. In this paper we critically examine four major formulations of the robot learning problem: inductive concept learning, explanationbased learning, reinforcement learning, and evolutionary learning. We describe some well-known examples of systems that fit under each formulation, and discuss their strengths and limitations. 1
|
4828
|
Genetic Algorithms
– Goldberg
- 1989
|
|
2489
|
Induction of Decision Trees
– Quinlan
- 1986
|
|
1364
|
A theory of the learnable
– Valiant
- 1984
|
|
939
|
Learning from Delayed Rewards
– Watkins
- 1989
|
|
693
|
Genetic Programming
– Koza
- 1992
|
|
471
|
Explanationbased generalization: a unifying view
– Mitchell, Keller, et al.
- 1986
|
|
377
|
Neuronlike adaptive elements that can solve difficult learning control problems
– Barto, Sutton, et al.
- 1983
|
|
375
|
Integrated Architectures for Learning, Planning and Reacting based on Approximate Dynamic Programming. Appeared
– Sutton
- 1990
|
|
331
|
Explanation-based learning: an alternative view
– Dejong, Mooney
- 1986
|
|
306
|
Automatic programming of behavior-based robots using reinforcement learning
– Mahadevan, Connell
- 1991
|
|
297
|
High resolution maps from wide angle sonar
– Moravec, Elfes
- 1985
|
|
287
|
Practical issues in temporal difference learning
– Tesauro
- 1992
|
|
252
|
Learning in Embedded Systems
– Kaelbling
- 1993
|
|
174
|
Reinforcement learning for robots using neural networks
– Lin
- 1993
|
|
123
|
Building Brains for Bodies
– Brooks, Stein
- 1994
|
|
119
|
Machine learning: a theoretical approach
– Natarajan
- 1991
|
|
118
|
Approximating optimal policies for partially observable stochastic domains. (unpublished manuscript
– Parr, Russell
- 1995
|
|
90
|
Map learning and high-speed navigation in RHINO
– Thrun, Bücken, et al.
|
|
83
|
Purposive behavior acquisition for a real robot by vision-based reinforcement learning
– Asada, Noda, et al.
- 1996
|
|
64
|
Passive distance learning for robot navigation
– Koenig, Simmons
- 1996
|
|
48
|
Learning one more thing
– Thrun, Mitchell
- 1994
|
|
47
|
editors. Robot Learning
– Connel, Mahadevan
- 1993
|
|
45
|
Enhancing transfer in reinforcement learning by building stochastic models of robot actions
– Mahadevan
- 1992
|
|
39
|
Alecsys and the AutonoMouse: Learning to Control a Real Robot by Distributed Classifier Systems
– Dorigo
- 1995
|
|
31
|
An evolutionary approach to learning in robots
– Grefenstette, Schultz
- 1994
|
|
31
|
Markov decision processes: Discrete dynamic programming
– Puterman
- 1994
|
|
17
|
Genetic Algorithms and Robotics
– Davidor
- 1991
|
|
16
|
Introduction to the special issue on learning autonomous robots
– Dorigo
- 1996
|
|
13
|
Neural network based autonomous navigation
– Pomerleau
- 1990
|
|
12
|
Certainty grids for mobile robots
– Moravec
- 1987
|
|
6
|
A case study of explanation-based control
– Dejong
- 1995
|
|
5
|
Learning policies for partially-observable environments: scaling up
– Littman, Cassandra, et al.
- 1995
|
|
5
|
Explanation-based learning: A comparison of symbolic and neural network approaches
– Mitchell, Thrun
- 1993
|
|
4
|
The Prospective student's Introduction to the Robot Learning Problem
– Nehmzow, Mitchell
- 1995
|
|
3
|
editors. Robot Learning. Number 2-3
– unknown authors
- 1996
|
|
1
|
Proceedings of the workshop on robot learning. Held in conjuction with
– Mahadevan
- 1994
|
