Search

In this paper, we investigate an approach for robots to extract the preferences of human observers, and combine them to generate new moves in order to improve robot dancing. Human preferences can be extracted even when a reward is given a few steps after a dance movement. With the feedback the robots perform more of what was preferred and less of what was not preferred. Human observers watch the robot generated dance movements and provide feedback in real time; then the robot learns the observers' preferences and creates new dance movements based on varying percentage of their preferences; and finally the observers rate the new robot's dancing. Experimental results show that the robot learns, using Interactive Reinforcement Learning, the expressed preferences of human observers and dance routines based on preferences of multiple observers are rated more highly.

Business organizations seldom work in isolation; effective interaction and cooperation between different organizations is essential for the success of any organization. In order to work together, organizations must have compatible workflows. This paper proposes a framework for automatically creating compatible workflows from high level requirements given by organizations that intend to work together. The framework uses intelligent agents [1] for the creation and execution of workflows. The proposed framework is different from existing systems as they focus on collaboration by modifying existing workflows. The proposed framework will enable organizations to avoid the time consuming task of creating a workflow and then ensuring that it is compatible with every organization it interacts with.

The signed directed graph (SDG) is the most commonly used type of model for automated hazard identification in chemical plants. Although SDG models are efficient in simulating the plant, they have some weaknesses, which are discussed here in relation to typical process industry examples. Ways to tackle these problems are suggested, and the view is taken that a state-based formalism is needed, to take account of the discrete components in the system, their connection together, and their behaviour over time. A strong representation for operations and actions is also needed, to make the models appropriate for modelling batch processes.

Industrial safety can be regarded as a major issue of industrial environments nowadays. This is why industries are currently spending huge amounts of resources to improve safety in all levels by reducing risks of causing damages to equipment, human injuries or even fatalities. This paper describes how Virtual Reality, Human Factors and Rule-based technologies are used in the framework of the VIRTHUALIS Integrated Project towards industrial training, safety management and accident investigation. The paper focuses mainly on the VR system specification and basic modules, while at the same time it presents the main system modules that synthesize the tool as a whole.

Driven by popular adoptions of workflow and requirements from the practice of virtual enterprise (VE), research in workflow interoperability is currently on the increase. Nonetheless, it is still in its early stage compared with the maturity of individual workflow technology. Some attempts have been tried, however results are not satisfactory especially in a VE context, where many of the partnerships are dynamic and temporary. Reasons include the rigidity and high initial coordination cost inherently associated with top-down modelling and enactment approaches. Therefore, this paper proposes a bottom-up and WfMS-independent approach towards cross-organisational workflow enactment, which is via progressive linking enabled by run-time agents. This is expected to pave the way for further cross-organisational workflow needs.

The ISA (is-a) hierarchies are widely used in the classification and the representation of related objects. In terms of assessing similarity between two nodes, current distance approaches suffer from its nature that only parent-child relationships among nodes are captured in the hierarchy. This paper presents an idea of treating a hierarchy as a set rather than as a tree in the traditional view. The established set theory is applied to provide the foundation where the relations between nodes can be mathematically specified and results in a more powerful and logical assessment of similarity between nodes.

Operating Procedure Synthesis (OPS) has been used to produce operating procedures for chemical plants. However, the application of AI planning to this domain has rarely been considered, and when it has the scope of the system used has limited it to solving ‘toy’ problems.