This paper presents a new anytime incremental search algorithm, AD*-Cut. AD*-Cut is based on two algorithms, namely, Anytime Repairing A* (ARA*) and the novel incremental search algorithm, D* Extra Lite. D* Extra Lite reinitializes (cuts) entire search-tree branches that have been affected by changes in an environment, and D* Extra Lite appears to be quicker than the reinitialization during the search utilized by the popular incremental search algorithm, D* Lite. The search-tree branch cutting is a simple and robust technique that can be easily applied to ARA*. Consequently, AD*-Cut extends D* Extra Lite in the same manner, as the state-of-the-art Anytime D* (AD*) algorithm extends D* Lite. The benchmark results suggest that AD*-Cut is quicker and achieves shorter paths than AD* when used for path planning on 3D state-lattices (a 2D position with rotation).

Motion-planning problems, such as manipulation in cluttered environments, often require a collision-free shortest path to be computed quickly given a roadmap graph. Typically, the computational cost of evaluating whether an edge of the roadmap graph is collision-free dominates the running time of search algorithms. Algorithms such as Lazy Weighted A* (LWA*) and LazySP have been proposed to reduce the number of edge evaluations by employing a lazy lookahead (one-step lookahead and infinite-step lookahead, respectively). However, this comes at the expense of additional graph operations: the larger the lookahead, the more the graph operations that are typically required. We propose Lazy Receding-Horizon A* (LRA*) to minimize the total planning time by balancing edge evaluations and graph operations. Endowed with a lazy lookahead, LRA* represents a family of lazy shortest-path graph-search algorithms that generalizes LWA* and LazySP. We analyze the theoretic properties of LRA* and demonstrate empirically that, in many cases, to minimize the total planning time, the algorithm requires an intermediate lazy lookahead. Namely, using an intermediate lazy lookahead, our algorithm outperforms both LWA* and LazySP. These experiments span simulated random worlds in R 2 and R 4, and manipulation problems using a 7-DOF manipulator.

We present a new approach to learning for planning, where knowledge acquired while solving a given set of planning problems is used to plan faster in related, but new problem instances. We show that a deep neural network can be used to learn and represent a generalized reactive policy (GRP) that maps a problem instance and a state to an action, and that the learned GRPs efficiently solve large classes of challenging problem instances. In contrast to prior efforts in this direction, our approach significantly reduces the dependence of learning on handcrafted domain knowledge or feature selection. Instead, the GRP is trained from scratch using a set of successful execution traces. We show that our approach can also be used to automatically learn a heuristic function that can be used in directed search algorithms. We evaluate our approach using an extensive suite of experiments on two challenging planning problem domains and show that our approach facilitates learning complex decision making policies and powerful heuristic functions with minimal human input.

Compilation techniques in planning reformulate a problem into an alternative encoding for which efficient, off-the-shelf solvers are available. In this work, we present a novel mixed-integer linear programming (MILP) compilation for cost-optimal numeric planning with instantaneous actions. While recent works on the problem are restricted to actions that modify variables present in simple numeric conditions, our MILP formulation, in addition, handles linear conditions and linear action effects on numeric state variables. Such problems are particularly challenging due to the state-dependency of the action effects. Experiments show that our approach, in addition to being the state of the art for the more general problem class, is competitive with heuristic search-based planners on domains with only simple numeric conditions.

Permutation flowshop scheduling problem (PFSP) is a classical combinatorial optimisation problem. There exist variants of PFSP to capture different realistic scenarios, but significant modelling gaps still remain with respect to real-world industrial applications such as the cider production line. In this paper, we propose a new PFSP variant that adequately models both overlapable sequence-dependent setup times (SDST) and mixed blocking constraints. We propose a computational model for makespan minimisation of the new PFSP variant and show that the time complexity is NP Hard. We then develop a constraint-guided local search algorithm that uses a new intensifying restart technique along with variable neighbourhood descent and greedy selection. The experimental study indicates that the proposed algorithm, on a set of wellknown benchmark instances, significantly outperforms the state-of-the-art search algorithms for PFSP.

This paper presents two prototypical epistemic forward planners, called EFP and PG-EFP, for generating plans in multi-agent environments. These planners differ from recently developed epistemic planners in that they can deal with unlimited nested beliefs, common knowledge, and capable of generating plans with both knowledge and belief goals. EFP is simply a breadth first search planner while PG-EFP is a heuristic search based system. To generate heuristics in PG-EFP, the paper introduces the notion of an epistemic planning graph. The paper includes an evaluation of the planners using benchmarks collected from the literature and discusses the issues that affect their scalability and efficiency, thus identifies potentially directions for future work. It also includes experimental evaluation that proves the usefulness of epistemic planning graphs.

We present several new algorithms for bidirectional best-first search that employ a front-to-front strategy of estimating distances from newly-generated frontier nodes in one search direction to existing frontier nodes in the other search direction, rather than estimating distances to terminal nodes in both searches. Unlike previous front-to-front strategies that use a shared priority queue to manage both frontiers, we use a separate data structure for each search, and choose that data structure to minimize the amount of computational effort required by the best-first search algorithm it supports.

Planning under partial observability in nondeterministic domains is a very significant and challenging problem, which requires dealing with uncertainty together with and-or search. In this paper, we propose a new algorithm for tackling this problem, able to generate conditional plans that are guaranteed to achieve the goal despite of the uncertainty in the initial condition and the uncertain effects of actions. The proposed algorithm combines heuristic search in the and-or space of beliefs with symbolic BDD-based techniques, and is fully amenable to the use of selection functions. The experimental evaluation shows that heuristic-symbolic search may behave much better than state-of-the-art search algorithms, based on a depth-first search (DFS) style, on several domains.

We present a random generator of partially complete round robin timetables that produces exclusively satisfiable instances, and provide experimental evidence that there is an easy-hard-easy pattern in the computational difficulty of completing partially complete timetables as the ratio of the number of removed entries to the total number of entries of the timetable is varied. Timetables in the hard region provide a suitable test-bed for evaluating and fine-tuning local search algorithms.

Local search algorithms are among the most effective approaches for solving the JSP, yet we have little understanding of why these algorithm work so well, and under what conditions. We develop descriptive cost models of local search for the job-shop scheduling problem (JSP), borrowing from the models developed for MAX-SAT. We show that several factors known to influence the difficulty of local search in MAX-SAT directly carry over to the general JSP, including the number of optimal solutions, backbone size, the distance between initial solutions and the nearest optimal solution, and an analog of backbone robustness. However, these same factors only weakly influence local search cost in JSPs with workflow, which possess structured constraints. While the factors present in the MAX-SAT cost models provide an accurate description of local search cost in the general JSP, our results for workflow JSPs raise concerns regarding the applicability of cost models derived using random problems to those exhibiting specific structure.