Artificial intelligence and robotic construction plant-the `smart artisan´

Robots and automated systems are now well established in the manufacturing industry where the operating environment can in general be modified and adapted to their requirements. As yet, however, these same technologies have had little impact in the unstructured, hostile and highly variable environment of a construction site. In order to function and survive in a site environment, construction robots will have to be highly adaptive, responding rapidly to changes in their operating conditions by making a series of strategic and tactical decisions about their mode of operation. This may be illustrated by considering the requirements for an automated and robotic excavator. The development of a hierarchical, Al based control system for robotic excavation has been the objective of the Lancaster Construction Robotics Group. Referred to as the `smart artisan´ because of its function and application this system has been implemented as the Lancaster University Computerised Intelligent Excavator (LUCIE) project on a JCB801 tracked excavator and has proved capable of autonomously digging a trench to a controlled depth in a variety of ground types and conditions and of overcoming and removing obstacles such as boulders without operator intervention