Decision-Theoretic Planning with Concurrent Temporally Extended Actions
This addresses planning efficiency in robotics or AI systems by enabling overlapping actions, but it is incremental as it builds on existing options and SMDP methods.
The paper tackles planning under uncertainty with concurrent temporally extended actions by combining the options framework with factored state spaces, showing that this defines a semi-Markov decision process and enabling the use of SMDP algorithms. In a navigation task experiment, concurrent options resulted in faster plans compared to sequential execution.
We investigate a model for planning under uncertainty with temporallyextended actions, where multiple actions can be taken concurrently at each decision epoch. Our model is based on the options framework, and combines it with factored state space models,where the set of options can be partitioned into classes that affectdisjoint state variables. We show that the set of decisionepochs for concurrent options defines a semi-Markov decisionprocess, if the underlying temporally extended actions being parallelized arerestricted to Markov options. This property allows us to use SMDPalgorithms for computing the value function over concurrentoptions. The concurrent options model allows overlapping execution ofoptions in order to achieve higher performance or in order to performa complex task. We describe a simple experiment using a navigationtask which illustrates how concurrent options results in a faster planwhen compared to the case when only one option is taken at a time.