Modern myoelectric prostheses remain difficult to control, particularly during rehabilitation, leading to high abandonment rates in favor of static devices. This highlights the need for advanced controllers that can automate some motions. This study presents an end-to-end framework coupling deep reinforcement learning with augmented reality (AR) for prosthetic actuation. A 14-degree-of-freedom hand was modeled in Blender and deployed in Unity. Two reinforcement learning agents were trained with distinct reward functions for a grasping task: (i) a discrete, Booleann reward with contact penalties and (ii) a continuous distance-based reward between joints and the target object. Each agent trained for 3 × 10⁷ timesteps at 50 Hz. The Booleann reward function performed poorly by entropy and convergence metrics, while the continuous reward function achieved success. The trained agent using the continuous reward was integrated into a dynamic AR scene, where a user controlled the prosthesis via a myoelectric armband while the grasping motion was actuated automatically. This framework demonstrates potential for assisting patients by automating certain movements to reduce initial control difficulty and improve rehabilitation outcomes.

As advanced reactors are nearing end-stage development as part of the growing effort to resolve society’s energy crisis, the interest in reprocessing used nuclear fuel is increasing. Pyroprocessing is a tested reprocessing technology requiring additional study before it may be considered for commercial viability. The Resonance Absorption Densitometry for Materials Assay Security Safeguards (RADMASS) program aims to study a non-destructive assay (NDA) method and create a safeguards model that may aid in the commercial development of pyroprocessing. Measurement uncertainties and time for input accountancy measurements are incorporated into the safeguards model, which is used to inform the development of a pyroprocessing facility model. Preliminary safeguards modeling results show that the magnitude of input accountancy uncertainty is of critical importance, with other factors vary. The initial findings from the safeguards model generated measurement averages that match what is expected for a real facility. Results from modeling assist in the understanding that as throughput size of a facility increases, the standard error of inventory difference (SEID) and inventory difference (ID) values also increase. While these results represent great progress toward the commercial viability of pyroprocessing, there is still more to be done before commercialization may proceed.

Augmented Reality (AR) with the assistance of myoelectric sensors has been shown to be a powerful tool in virtual rehabilitation for trans-radial arm amputees. Current methods utilize a pre-dimensioned and fabricated virtual prosthetic inside of the AR space to allow users to practice controlling a virtual prosthetic. This practice is in preparation for daily activities while using the prosthetic device. Understanding the potential that AR has in shaping the rehabilitation process, we aim to investigate whether this potential can carry over in the fitting process of the prosthetic device, i.e., the dimensioning of the prosthesis and its accuracy down to the millimeter level. To accurately obtain the dimensions necessary for the creation of an upper-extremity prosthetic an AR application is developed and launched through Microsoft's HoloLens 2 headset. The proposed method uses grabbable virtual markers that assist in obtaining the measurements and dimensions of a virtual prosthetic. The sensitivity and accuracy of the application is analyzed across usage from multiple participants with the mean average error generally ranging from 1–2 mm.