A hybrid POMDP-BDI agent architecture with online stochastic planning and plan caching.
Gavin RensDeshendran MoodleyPublished in: Cogn. Syst. Res. (2017)
Keyphrases
- agent architecture
- belief desire intention
- planning problems
- plan generation
- ai planning
- planning process
- travel planning
- planning domains
- intelligent agents
- stochastic domains
- plan execution
- model free reinforcement learning
- multi agent systems
- bdi architecture
- partially observable markov decision process
- partially observable
- belief space
- planning systems
- plan recognition
- software agents
- bdi agents
- decision theoretic
- state space
- partially observable markov decision processes
- layered architecture
- reinforcement learning
- hybrid architecture
- agent based architecture
- classical planning
- mental attitudes
- cooperative
- query processing
- finite state
- optimal plans
- bdi model
- markov decision processes
- initial state
- agent platform
- belief state
- heuristic search
- dec pomdps
- decision problems
- multiagent systems
- multi agent