Over the past decades, robots have been extensively deployed in multiple industries. More recently, industrial robots have been taken out of their cages, being more present in dynamic environments, interacting in the close vicinity of human operators. Traditionally, robots have been mainly developed to perform pre-programmed tasks. However, some tasks are too complex or expensive to be performed by a robotic system alone. Examples are the handling of large sheet-like objects in the composite part production or plastic film wrapping industry. This work presents a hybrid wrench and vision reactive control approach towards the handling of large (non-)rigid materials. The presented approach fuses force-torque data and skeleton tracking to control a mobile manipulator in an intuitive manner, by using the intelligence of the operator as much as possible. Using this approach, tools such as path planning, or object modelling are not essential to obtain the results. The hybrid controller is subject to stability experiments where the controller responses are monitored when the mobile manipulator is subject to a step and sinusoidal function as input. Lastly, the overall approach is illustrated with a proof-of-concept task in which a flexible sheet is handled by a mobile manipulator and human operator together.
This study determines the minimal detection time (MDT) needed for successful localization of radioactive hot spots during nuclear decommissioning work. An automated XY stage, equipped with a CdZnTe (CZT) spectrometer, was used to identify and localize hot spots of 241Am, 137Cs, and 60Co in a 1.7 × 1.7-m area. The stage served as a preliminary test platform for the development of an automated robotic characterization platform [Autonomous Robotic platform for CHaractERization (ARCHER) robot]. The dependence of the MDT on the detector efficiency and background (BKG) level was examined. For low BKG environments, the MDT for 137Cs was 871 ms and resulted in an error of the source localization of 14.21 mm and an error of the activity of 6.85%. For elevated BKG levels, the MDT increased to 15 526 ms. The 137Cs source was localized with an error of 34.13 mm and an error of the source activity of −7.04%. The MDT determination method used here offers a valuable approach for decreasing total scanning times while avoiding missing the presence of hot spots.
3 February 2026 in Journal of Pharmaceutical Innovation
This paper presents a comprehensive optimisation and evaluation of an automated de-bagging process for aseptic manufacturing, where a double-bagged product must be transferred from a non-sterile to a highly sterile environment. The current system experiences a high failure rate, with a success rate of only \(\varvec{30~\%}\) in the initial de-bagging step, leading to significant product loss and production downtime. Through a series of empirical, mechanical, and push-out force-based optimisations, including improvements to the product guide rails, clamping mechanisms, and pusher bar operation the success rate of the de-bagging process was increased to \(\varvec{89~\%}\). Based on the results a set of generalised recommendations was formalised for the design of other de-baggers. Part of the suggested optimisation methods have since been integrated in a real de-bagger setup, which shows similar results to those achieved in the test setup.