The workshop 'AI-based Planning for Cyber-Physical Systems', which took place on February 26, 2024, as part of the 38th Annual AAAI Conference on Artificial Intelligence in Vancouver, Canada, brought together researchers to discuss recent advances in AI planning methods for Cyber-Physical Systems (CPS). CPS pose a major challenge due to their complexity and data-intensive nature, which often exceeds the capabilities of traditional planning algorithms. The workshop highlighted new approaches such as neuro-symbolic architectures, large language models (LLMs), deep reinforcement learning and advances in symbolic planning. These techniques are promising when it comes to managing the complexity of CPS and have potential for real-world applications.

This is the first work to look at the application of large language models (LLMs) for the purpose of model space edits in automated planning tasks. To set the stage for this sangam, we explore two different flavors of model space problems that have been studied in the AI planning literature and explore the effect of an LLM on those tasks. We empirically demonstrate how the performance of an LLM contrasts with combinatorial search (CS) - an approach that has been traditionally used to solve model space tasks in planning, both with the LLM in the role of a standalone model space reasoner as well as in the role of a statistical signal in concert with the CS approach as part of a two-stage process. Our experiments show promising results suggesting further forays of LLMs into the exciting world of model space reasoning for planning tasks in the future.

AI Planning, Machine Learning and Process Mining have so far developed into separate research fields. At the same time, many interesting concepts and insights have been gained at the intersection of these areas in recent years. For example, the behavior of future processes is now comprehensively predicted with the aid of Machine Learning. For the practical application of these findings, however, it is also necessary not only to know the expected course, but also to give recommendations and hints for the achievement of goals, i.e. to carry out comprehensive process planning. At the same time, an adequate integration of the aforementioned research fields is still lacking. In this article, we present a research project in which researchers from the AI and BPM field work jointly together. Therefore, we discuss the overall research problem, the relevant fields of research and our overall research framework to automatically derive process models from executional process data, derive subsequent planning problems and conduct automated planning in order to adaptively plan and execute business processes using real-time forecasts.

According to a recent McKinsey survey of executives, companies have pushed the time frame for digitizing many aspects of their business, from internal operations to supply chain and customer interactions, by three to four years. Digital products in those companies' portfolios have also shot some seven years ahead of where they had expected to be prior to the pandemic. The Great Resignation, skill shortages, supply chain disruptions, working from home, touchless customer experience, and agile process redesigns are paradigm shifts that businesses have rapidly needed to adapt to. But how do companies stay dynamic, resilient, and efficient in this new era? We believe that a big part of the solution may be found in digital employees, powered by automation and AI.

To the right, search space, where all states with the same initial-state distance (g) and estimated goal distance (h) are represented by a single binary decision diagram (to the left), and only those whose g h solution cost need to be considered. Daniel Fišer, Álvaro Torralba and Joerg Hoffmann won an outstanding paper runners-up award at AAAI 2022 for their paper Operator-potential heuristics for symbolic search. Here, Álvaro tells us more about the field of optical planning, their methodology, and how potential heuristics can be used in symbolic search with very positive results. At a very general level, the research is on automated planning. This is a sub-area of AI where we try to answer the question: what is the best way to act given our knowledge of the world?

We describe a Graphical Interface for Planning with Objects called GIPO that has been built to investigate and support the knowledge engineering process in the building of applied AI planning systems. GIPO embodies an object centred approach to planning domain modelling. There are two reasons for providing knowledge engineering support for AI planning: (i) to apply a planning system to a new domain to test the planning system itself (ii) to tackle the end-user problem for the engineer who might be a domain expert but need not necessarily have a specialist knowledge of AI planning. Our research is primarily aimed at developing a method and tools to meet the requirements of the latter case (ii), although the benefits can also be enjoyed by planning experts.

Motivated by the requirements of many real-life applications, recent research in AI planning has shown a growing interest in tackling problems that involve numeric constraints and complex optimization objectives. Applying Integer Programming (IP) to such domains seems to have a significant potential, since it can naturally accommodate their representational requirements. In this paper we explore the area of applying IP to AI planning in two different directions. First, we improve the domain-independent IP formulation of Vossen et al., by an extended exploitation of mutual exclusion relations between the operators, and other information derivable by state of the art domain analysis tools. This information may reduce the number of variables of an IP model and tighten its constraints. Second, we link IP methods to recent work in heuristic search for planning, by introducing a variant of {\tt FF}'s enforced hill-climbing algorithm that uses IP models as its underlying representation. In addition to extending the delete lists heuristic to parallel planning and the more expressive language of IP, we also introduce a new heuristic based on the linear relaxation.

In recent years, we have witnessed tremendous progress in deep reinforcement learning (RL) for tasks such as Go, Chess, video games, and robot control. Nevertheless, other combinatorial domains, such as AI planning, still pose considerable challenges for RL approaches. The key difficulty in those domains is that a positive reward signal becomes {\em exponentially rare} as the minimal solution length increases. So, an RL approach loses its training signal. There has been promising recent progress by using a curriculum-driven learning approach that is designed to solve a single hard instance. We present a novel {\em automated} curriculum approach that dynamically selects from a pool of unlabeled training instances of varying task complexity guided by our {\em difficulty quantum momentum} strategy. We show how the smoothness of the task hardness impacts the final learning results. In particular, as the size of the instance pool increases, the ``hardness gap'' decreases, which facilitates a smoother automated curriculum based learning process. Our automated curriculum approach dramatically improves upon the previous approaches. We show our results on Sokoban, which is a traditional PSPACE-complete planning problem and presents a great challenge even for specialized solvers. Our RL agent can solve hard instances that are far out of reach for any previous state-of-the-art Sokoban solver. In particular, our approach can uncover plans that require hundreds of steps, while the best previous search methods would take many years of computing time to solve such instances. In addition, we show that we can further boost the RL performance with an intricate coupling of our automated curriculum approach with a curiosity-driven search strategy and a graph neural net representation.

This is yet another approach to improve the expressiveness and complexity of the classical planning approaches. In this technique what we do is convert our classical planning problems representation into a well-known representation called Propositional Satisfiability Problem, also called Boolean Satisfiability Problem, or simply SAT. If we can convert it to a well-known problem representation, we can use existing algorithms to solve the problems. If you are not familiar with classical planning approaches, please read the Introduction section in my previous post linked below. In this post, I'll explain the concept by going through examples step-by-step.