The nuclear industry has some of the most extreme environments in the world, with radiation levels and extremely harsh conditions restraining human access to many facilities. One method for enabling minimal human exposure to hazards under these conditions is through the use of gloveboxes that are sealed volumes with controlled access for performing handling. While gloveboxes allow operators to perform complex handling tasks, they put operators at considerable risk from breaking the confinement and, historically, serious examples including punctured gloves leading to lifetime doses have occurred. To date, robotic systems have had relatively little impact on the industry, even though it is clear that they offer major opportunities for improving productivity and significantly reducing risks to human health. This work presents the challenges of robotic and AI solutions for nuclear gloveboxes, and introduces a step forward for bringing cutting-edge technology to gloveboxes. The problem statement and challenges are highlighted and then an integrated demonstrator is proposed for robotic handling in nuclear gloveboxes for nuclear material handling. The proposed approach spans from tele-manipulation to shared autonomy, computer vision solutions for robotic manipulation to machine learning solutions for condition monitoring.

Control of systems of automated guided vehicles involves action planning at many levels. For efficient control of these systems, accurate estimation of cost parameters (speed, energy, task completion performance, et cetera is required. These parameters change along time, particularly in battery-operated robots, which are very sensitive to battery level variations. This work addresses the problem of on-line cost parameter identification and estimation for proper control decisions of the individual mobile robots and for the system as a whole. Several filtering and estimation methods have been investigated with respect to travelling times, which are dramatically affected by battery charges and condition of facility's floors, among other factors. Results show that these parameters depend on the robot, the route and the moment, so they are linked to a particular robot, a region of the floor and a time period (or to a battery level). Moreover, differences with static, pre-runtime travelling time computations, either heuristically or by characterization of real robots, are large enough to affect to system's performance and overall productivity and efficiency.

No abstract

No abstract

The costs incurred in a mobile robot (MR) change due to change in physical and environmental factors. Usually, there are two approaches to consider these costs, either explicitly modelling these different factors to calculate the cost or consider heuristics costs. First approach is lengthy and cumbersome and requires a new model for every new factor. Heuristics cost cannot account for the change in cost due to change in state. This work proposes a new method to compute these costs, without the need of explicitly modelling the factors. The identified cost is modelled in a bi-linear state-space form where the change of costs is formed due to the change of these states. This eliminates the need to model all factors to derive the cost for every robot. In context of transportation, the travel time is identified as the key parameter to understand costs of traversing paths to carry material. The necessity to identify and estimate these travel times is proved by using them in route planning. The paths are computed constantly and average of total path costs of these paths are compared with that of paths obtained by heuristics costs. The results show that average total path costs of paths obtained through on-line estimated travel times are 15% less that of paths obtained by heuristics costs.