Towards optimised robotic assembly

Assembly by robot is destined to have a major international impact over the next few years, yet the optimal design, analysis and economic evaluation of a potential robotic assembly application is currently extremely difficult. On looking at the current state-of-the-art in each stage of the hierarchy which a prospective robotic assembly project must pass through, it becomes clear that the whole procedure is closely analogous to attempting to paint an 'ideal' picture--both from the point of view of tasks which must be performed and also the level of complexity involved. Unfortunately, unlike with painting, there is not yet any wealth of experience on which roboticists can draw in their quest for 'ideal' assembly. Nevertheless, work at Imperial College aims at systematically uncovering some of the underlying principles of robotic assembly which if left to discovery through trial and error might take years--years which industry simply does not have.