Learning action models for the improved execution of navigation plans

Robotics and Autonomous Systems

Volumn 38, Issue 3-4, 2002, Pages 137-148

  Belker, Thorsten ^a   Beetz, Michael ^b   Cremers, Armin B ^a  

^a UNIVERSITY OF BONN   (Germany)

^b TECHNICAL UNIVERSITY OF MUNICH   (Germany)

Author keywords

Action selection; Regression trees; Robot learning; Robot navigation

Indexed keywords

COMPUTER SIMULATION; LEARNING ALGORITHMS; LEARNING SYSTEMS; MARKOV PROCESSES; NAVIGATION SYSTEMS; NEURAL NETWORKS;

NAVIGATION PLANNING;

ROBOTS;

EID: 0037204477     PISSN: 09218890     EISSN: None     Source Type: Journal    
DOI: 10.1016/S0921-8890(02)00163-X     Document Type: Article

References (25)

1. Acting under uncertainty: Discrete Bayesian models for mobile-robot navigation
2. Probabilistic robot navigation in partially observable environments
3. AI-based mobile robots: Case studies of successful robot systems
4. The dynamic window approach to collision avoidance
5. The curvature-velocity method for local obstacle avoidance
6. Environment and task adaptation for robotic agents
7. Integrating active localization into high-level control systems
8. Planning and acting in partially observable stochastic domains
9. Markov localization for mobile robots in dynamic environments
10. Active mobile robot localization
11. A gradient method for real-time robot control
12. Unsupervised learning of probabilistic models for robot navigation
13. An approach to learning mobile robot navigation
14. Continuous case-based reasoning
15. Multi-agent Q-learning and regression trees for automated pricing decisions
16. Efficient value function approximation using regression trees
17. Using regression trees to learn action models
18. Team-partitioned, opaque-transition reinforced learning