Handling Mitigating Circumstances for Electronic Contracts

Electronic contracts are a means of representing agreed responsibilities and expected behaviour of autonomous agents acting on behalf of businesses. They can be used to regulate behaviour by providing negative consequences, penalties, where the responsibilities and expectations are not met, i.e. the contract is violated. However, long-term business relationships require some flexibility in the face of circumstances that do not conform to the assumptions of the contract, that is, mitigating circumstances. In this paper, we describe how contract parties can represent and enact policies on mitigating circumstances. As part of this, we require records of what has occurred within the system leading up to a violation: the provenance of the violation. We therefore bring together contract-based and provenance systems to solve the issue of mitigating circumstances.

Modelling and Administration of Contract-Based Systems

Mirroring the paper versions exchanged between businesses today, electronic contracts offer the possibility of dynamic, automatic creation and enforcement of restrictions and compulsions on agent behaviour that are designed to ensure business objectives are met. However, where there are many contracts within a particular application, it can be difficult to determine whether the system can reliably fulfil them all; computer-parsable electronic contracts may allow such verification to be automated. In this paper, we describe a conceptual framework and architecture specification in which normative business contracts can be electronically represented, verified, established, renewed, etc. In particular, we aim to allow systems containing multiple contracts to be checked for conflicts and violations of business objectives. We illustrate the framework and architecture with an aerospace example.

A Model of Normative Power

A power describes the ability of an agent to act in some way. While this notion of power is critical in the context of organisational dynamics, and has been studied by others in this light, it must be constrained so as to be useful in any practical application. In particular, we are concerned with how power may be used by agents to govern the imposition and management of norms, and how agents may dynamically assign norms to other agents within a multi-agent system. We approach the problem by defining a syntax and semantics for powers governing the creation, deletion, or modification of norms within a system, which we refer to as normative powers. We then extend this basic model to accommodate more general powers that can modify other powers within the system, and describe how agents playing certain roles are able to apply powers, changing the system’s norms, and also the powers themselves. We examine how the powers found within a system may change as the status of norms change, and show how standard norm modification operations — such as the derogation, annulment and modification of norms— may be represented within our system.

BDI4JADE: a BDI layer on top of JADE

Several agent platforms that implement the belief-desire-intention (BDI) architecture have been proposed. Even though most of them are implemented based on existing general purpose programming languages, e.g. Java, agents are either programmed in a new programming language or Domain-specific Language expressed in XML. As a consequence, this prevents the use of advanced features of the underlying programming language and the integration with existing libraries and frameworks, which are essential for the development of enterprise applications. Due to these limitations of BDI agent platforms, we have implemented the BDI4JADE, which is presented in this paper. It is implemented as a BDI layer on top of JADE, a well accepted agent platform.

Extending Agent Languages for Autonomy

BDI agent languages provide a useful abstraction for complex systems comprised of interactive autonomous entities, but they have been used mostly in the context of single agents with a static plan library of behaviours invoked reactively. These languages provide a theoretically sound basis for agent design but are very limited in providing direct support for autonomy and societal cooperation needed for large scale systems. Some techniques for autonomy and cooperation have been explored in the past in ad hoc implementations, but not incorporated in any agent language. In order to address these shortcomings we extend the well known AgentSpeak(L) BDI agent language to include behaviour generation through planning, declarative goals and motivated goal adoption. We also develop a language-specific multiagent cooperation scheme and, to address potential problems arising from autonomy in a multiagent system, we extend our agents with a mechanism for norm processing leveraging existing theoretical work. These extensions allow for greater autonomy in the resulting systems, enabling them to synthesise new behaviours at runtime and to cooperate in non-scripted patterns.