ARTS, an implementation of the real-time BDI agent architecture. ARTS is an agent programming framework for agents with soft real-time guarantees; an ARTS agent will attempt to achieve as many high priority tasks by their specified deadlines as possible. The syntax and execution semantics of ARTS is based that of PRS-CL and JAM, augmented with information about deadlines, priorities, and durations, and changes to the interpreter to implement time bounded priority driven plan selection and deadline monotonic intention scheduling. ARTS is implemented in Java, and the current prototype implementation includes the core language described below, and implementations of some basic primitive actions. Additional user-defined primitive actions can be added using a Java API.
An ARTS agent consists of five main components: a database, a goal stack, a plan library, an intention structure, and an interpreter. Changes to the agent's environment or posting of new goals invokes reasoning to search for plans that might be applied to the current situation. The ARTS interpreter selects one plan from the list of applicable plans, schedules it, and executes a step of the most urgent intention in the computed schedule.
• AgentSpeak(RT): Agent System with deadlines and priorities
AgentSpeak(RT) is a real-time BDI agent programming language based on AgentSpeak(L). AgentSpeak(RT) extends AgentSpeak intentions with deadlines which specify the time by which the agent should respond to an event, and priorities which specify the relative importance of responding to a particular event. The AgentSpeak(RT) interpreter commits to a priority-maximal set of intentions: a set of intentions which is maximally feasible while preferring higher priority intentions. We prove some properties of the language, such as guaranteed reactivity delay of the AgentSpeak(RT) interpreter and probabilistic guarantees of successful execution of intentions by their deadlines.