Research Projects

Transradial amputations (loss of the forearm below the elbow) account for a significant number of upper limb amputations worldwide. People living with transradial limb loss face many daily challenges, from basic grasping and holding to a wide range of manipulation tasks, highlighting the limited prosthetic options for restoring higher levels of function and dexterity. While myoelectric and body-powered prosthetic hands are becoming more advanced, there remains a noticeable gap between what these devices can do and the complex needs of real-world use. Artificial hand development necessitates an extensive iterative process to fine-tune the actuator selection, parameter tuning, and design variables, and it also varies from user to user because each prosthetic user has unique muscle characteristics and a residual limb. Additionally, the development cycle necessitates trials with prosthetic users, which are typically time-consuming and logistically demanding, requiring ethical approval and subject recruitment. An alternative that avoids these shortcomings is offered by virtual simulation. Human movement simulation tools like OpenSim and AnyBody have been successful in musculoskeletal (MSKM) modelling. Nonetheless, to simulate the complex physical interactions between wearable robotic systems and humans, often multiple simulation platforms need to be coupled. Early research has employed musculoskeletal simulation platforms, including the Rajagopal full-body model, to simulate human gait. Optimizing and designing these Prostheses, nevertheless, requires predictive tools. Musculoskeletal simulation, particularly forward dynamics modelling in OpenSim, is being increasingly used for this purpose. Current research has tended to study muscle forces, joint loads, or secondary conditions like osteoarthritis and lumbar pain in amputees from passive prostheses. Other methods include optimal control-based predictive simulations to examine transtibial amputee gait regulation strategies. Yet, musculoskeletal models that freely incorporate prosthesis hand elements with explicit actuation are rare. Such simulations enable iterative assistive device design and control tuning, as well as provide a remedy to the lack of open-access tools for the design of powered prostheses. By simulating human tasks in a virtual manner, the approach circumvents ethical as well as logistical issues and speeds up the innovation cycle of high-level prosthetic technology. We propose a new simulation workflow combining OpenSim and MATLAB to simulate a passive and powered prosthetic hand. The method combines the lower-elbow musculoskeletal model with actuator dynamics and control laws. The hybrid simulation allows forward dynamics analysis without having to perform physical user task-based trials. The proposed model includes biologically realistic actuation, enabling the virtual testing of torque and joint kinematics across different task-specific conditions.