Architecture for Hybrid Robotic Behavior

Abstract

Abstract. Software architectures for agent technology and robots have been polarized between reactive architectures and architectures based on planning and reasoning. Although hybrid architectures have been shown to offer benefits from both, these seem complicated to integrate. In this paper we integrate the reactive nature of finite state machines and the reasoning capabilities of non-monotonic logics to produce intelligent autonomous robots. In particular, we demonstrate this with a robotic poker player. The robotic player integrates vision, sound recognition, motion control and the reasoning to perform competitively as a player in a complex game with incomplete information.

High Performance Relaying of C++11 Objects across Processes and Logic-Labeled Finite-State Machines

“…Thus, it is possible to include an entire reasoning system, for example using Prolog and invoke it from C++11 using standard APIs. This approach was used for poker hands [2] and to build a poker playing robot. We have found DPL, a common sense non-monotonic logic, very useful for declarative aspects.…”

“…Our gusimplewhiteboard has proven a very efficient and effective communication infrastructure of objects defined by fully fleshed C++11 classes. In combination with clfsm, they provide a very flexible control architecture [2] that minimises concurrency concerns and has facilitated the rapid development of complex, high-level behaviours through composition of modules and llfsms. Moreover, C++11's static type system enables far more secure software development.…”

Visual-Trace Simulation of Concurrent Finite-State Machines for Validation and Model-Checking of Complex Behaviour

“…Our FSMs use simple C statements 1 for states, and expressions for transitions (including function calls and evoking expert system evaluation of non-monotonic logic). The communication medium between modules of the software is based on a whiteboard architecture and details have been provided elsewhere [3].…”

A Defeasible Logic for Clauses

“…A key feature of these defeasible logics is that they all have efficient easily implementable deduction algorithms [10,20,22]. Indeed defeasible logics have been used in an expert system, for learning and planning [22], in a robotic dog which plays soccer [4,5], in a robotic poker player [6], to improve the accuracy of radio frequency identification [11], to model the behaviour of autonomous robots [9], and to facilitate the encoding of software requirements [8] so they can be automatically translated into a programming language [7]. Defeasible logics have been advocated for various applications including modelling regulations and business rules [1], agent negotiations [13], the semantic web [2,3,25], modelling agents [16], modelling intentions [15], modelling dynamic resource allocation [17], modelling contracts [12], legal reasoning [18], modelling deadlines [14], and modelling dialogue games [24].…”