Automatic behavior composition synthesis

Abstract

The behavior composition problem amounts to realizing a virtual desired module (e.g., a surveillance agent system) by suitably coordinating (and re-purposing) the execution of a set of available modules (e.g., a video camera, vacuum cleaner, a robot, etc.). In particular, we investigate techniques to synthesize a controller implementing a fully controllable target behavior by suitably coordinating available partially controllable behaviors that are to execute within a shared, fully observable, but partially predictable (i.e., non-deterministic), environment. Both behaviors and environment are represented as arbitrary finite state transition systems. The technique we propose is directly based on the idea that the controller job is to coordinate the concurrent execution of the available behaviors so as to "mimic" the target behavior. To this end, we exploit a variant of the formal notion of simulation to formally capture the notion of "mimicking", and we show that the technique proposed is sound and complete, optimal with respect to computational complexity, and robust for different kind of system failures. In addition, we demonstrate that the technique is well suited for highly efficient implementation based on synthesis by model checking technologies, by relating the problem to that of finding a winning strategy in a special safety game and explaining how to actually solve it using an existing verification tool. © 2013 Elsevier B.V. All rights reserved

Supervisory Control for Behavior Composition

“…The behavior composition problem has been recently much studied in the web-services and AI literature [4,6,11,26,15]. The problem amounts to synthesising a controller that is able to "realize" (i.e., implement) a desired, but nonexistent, target behavior module, by suitably coordinating a set of available behavior modules.…”

“…The problem amounts to synthesising a controller that is able to "realize" (i.e., implement) a desired, but nonexistent, target behavior module, by suitably coordinating a set of available behavior modules. In what follows, we mostly follow the model detailed in [11]. 1 Generally speaking, behaviors represent the operational logic of a device or program, and they are modeled using, possibly nondeterministic, finite transition systems.…”

“…Second, computational properties for supervisor synthesis have been substantially studied and tools for supervisor synthesis are available, including TCT/STCT [31], GRAIL [22], DESUMA [23], and SUPREM-ICA [18]. Thus, we can apply very different techniques for solving the composition problem than those already available within the AI literature (e.g., PDL satisfiability [10], direct search [26], LTL/ATL synthesis [15,12], and computation of special kind of simulation relations [11,6]). Finally, once linked, we expect cross-fertilization between AI-style composition and DES supervisory theory.…”

Verification and Synthesis in Description Logic Based Dynamic Systems

“…Here the system is the orchestrator, and the environment is formed by the target and the available behaviors. The goal of the system is to maintain over time the ability of delegating requested actions [18]. Several other sophisticated forms of synthesis and planning can be captured through 2GS, see, e.g., [17].…”

“…We recall that µ-calculus subsumes virtually all logics used in verification, including LTL, CTL and CTL * . Adversarial synthesis for µ-calculus captures a wide variety of synthesis problems, including conditional planning with full-observability [22], behaviour composition [18], and several other sophisticated forms of synthesis and planning [17].…”