We propose a model, Lexicographic Partially Observable Markov Decision Process (LPOMDP), which extends POMDPs with lexicographic preferences over multiple value functions. It allows for slack--slightly less-than-optimal values--for higher-priority preferences to facilitate improvement in lower-priority value functions. Many real life situations are naturally captured by LPOMDPs with slack. We consider a semi-autonomous driving scenario in which time spent on the road is minimized, while maximizing time spent driving autonomously. We propose two solutions to LPOMDPs--Lexicographic Value Iteration (LVI) and Lexicographic Point-Based Value Iteration (LPBVI), establishing convergence results and correctness within strong slack bounds. We test the algorithms using real-world road data provided by Open Street Map (OSM) within 10 major cities. Finally, we present GPU-based optimizations for point-based solvers, demonstrating that their application enables us to quickly solve vastly larger LPOMDPs and other variations of POMDPs.

Pruning techniques based on partial order reduction and symmetry elimination have recently found increasing attention for optimal planning. Although these techniques appear to be rather different, they base their pruning decisions on similar ideas from a high level perspective. In this paper, we propose safe integrations of partial order reduction and symmetry elimination for cost-optimal classical planning. We show that previously proposed symmetry-based search algorithms can safely be applied with strong stubborn sets. In addition, we derive the notion of symmetrical strong stubborn sets as a more tightly integrated concept. Our experiments show the potential of our approaches.

The development of domain-independent planners This modular approach also supports the use of reformulation within the AI Planning community is leading to and configuration techniques which can automatically "off the shelf" technology that can be used in a reformulate, re-represent or tune the domain model and/or wide range of applications. Moreover, it allows a problem description in order to increase the efficiency of modular approach - in which planners and domain a planner and increase the scope of problems solved. The knowledge are modules of larger software applications idea is to make these techniques to some degree independent - that facilitates substitutions or improvements of domain and planner (that is, applicable to a range of individual modules without changing the of domains and planning engine technologies), and use them rest of the system. This approach also supports the to form a wrapper around a planner, improving its overall use of reformulation and configuration techniques, performance for the domain to which it is applied. Types which transform how a model is represented in order of reformulation include macro-learning [Botea et al., 2005; to improve the efficiency of plan generation. Newton et al., 2007], action schema splitting [Areces et al., In this paper, we investigate how the performance 2014] and entanglements [Chrpa and McCluskey, 2012]: here of planners is affected by domain model configuration.

Linear temporal logic (LTL) is an expressive language that allows specifying temporally extended goals and preferences. A general approach to dealing with general LTL properties in planning is by ``compiling them away''; i.e., in a pre-processing phase, all LTL formulas are converted into simple, non-temporal formulas that can be evaluated in a planning state. This is accomplished by first generating a finite-state automaton for the formula, and then by introducing new fluents that are used to capture all possible runs of the automaton. Unfortunately, current translation approaches are worst-case exponential on the size of the LTL formula. In this paper, we present a polynomial approach to compiling away LTL goals. Our method relies on the exploitation of alternating automata. Since alternating automata are different from non-deterministic automata, our translation technique does not capture all possible runs in a planning state and thus is very different from previous approaches. We prove that our translation is sound and complete, and evaluate it empirically showing that it has strengths and weaknesses. Specifically, we find classes of formulas in which it seems to outperform significantly the current state of the art.

In optimal planning as heuristic search, admissible pruning techniques are paramount. One idea is dominance pruning, identifying states "better than" other states. Prior approaches are limited to simple dominance notions, like "more STRIPS facts true" or "higher resource supply". We apply simulation, well-known in model checking, to compute much more general dominance relations based on comparing transition behavior across states. We do so effectively by expressing state-space simulations through the composition of simulations on orthogonal projections. We show how simulation can be made more powerful by intertwining it with a notion of label dominance. Our experiments show substantial improvements across several IPC benchmark domains.

We consider planning problems for stochastic games with objectives specified by a branching-time logic, called probabilistic computation tree logic (PCTL). This problem has been shown to be undecidable if strategies with perfect recall, i.e., history-dependent, are considered. In this paper, we show that, if restricted to co-safe properties, a subset of PCTL properties capable to specify a wide range of properties in practice including reachability ones, the problem turns to be decidable, even when the class of general strategies is considered. We also give an algorithm for solving robust stochastic planning, where a winning strategy is tolerant to some perturbations of probabilities in the model. Our result indicates that satisfiability of co-safe PCTL is decidable as well.

Many sequential decision-making problems require an agent to reason about both multiple objectives and uncertainty regarding the environment's state. Such problems can be naturally modelled as multi-objective partially observable Markov decision processes (MOPOMDPs). We propose optimistic linear support with alpha reuse (OLSAR), which computes a bounded approximation of the optimal solution set for all possible weightings of the objectives. The main idea is to solve a series of scalarized single-objective POMDPs, each corresponding to a different weighting of the objectives. A key insight underlying OLSAR is that the policies and value functions produced when solving scalarized POMDPs in earlier iterations can be reused to more quickly solve scalarized POMDPs in later iterations. We show experimentally that OLSAR outperforms, both in terms of runtime and approximation quality, alternative methods and a variant of OLSAR that does not leverage reuse.

This work compares two actions' concurrency and co-occurrence employed in temporal modeling languages, one with a PDDL-style action modeling languages used by the AI planning community, exclusion mechanism, and another with an explicit and argue that they explain why MILP or SMT have notion of resources, and investigates their seemed unattractive. Specifically, we observe that PDDL 2.1 implications on constraint-based search. The first [Fox and Long, 2003] induces temporal gaps between consecutive mechanism forces temporal gaps in action schedules interdependent actions, and these gaps often induce and have a high performance penalty. The second twice the number of steps in the plans than what is necessary, mechanism avoids the gaps, with dramatically with strong negative performance implications. The gaps are improved performance.

Real-time strategy (RTS) games are hard from an AI point of view because they have enormous state spaces, combinatorial branching factors, allow simultaneous and durative actions, and players have very little time to choose actions. For these reasons, standard game tree search methods such as alpha- beta search or Monte Carlo Tree Search (MCTS) are not sufficient by themselves to handle these games. This paper presents an alternative approach called Adversarial Hierarchical Task Network (AHTN) planning that combines ideas from game tree search with HTN planning. We present the basic algorithm, relate it to existing adversarial hierarchical planning methods, and present new extensions for simultaneous and durative actions to handle RTS games. We also present empirical results for the μRTS game, comparing it to other state of the art search algorithms for RTS games.

Nowadays, multiagent planning under uncertainty scales to tens or even hundreds of agents. However, current methods either are restricted to problems with factored value functions, or provide solutions without any guarantees on quality. Methods in the former category typically build on heuristic search using upper bounds on the value function. Unfortunately, no techniques exist to compute such upper bounds for problems with non-factored value functions, which would additionally allow for meaningful benchmarking of methods of the latter category. To mitigate this problem, this paper introduces a family of influence-optimistic upper bounds for factored Dec-POMDPs without factored value functions. We demonstrate how we can achieve firm quality guarantees for problems with hundreds of agents.