Control of a Wheelchair-Mounted 6DOF Assistive Robot With Chin and Finger Joysticks

Abstract

Throughout the last decade, many assistive robots for people with disabilities have been developed; however, researchers have not fully utilized these robotic technologies to entirely create independent living conditions for people with disabilities, particularly in relation to activities of daily living (ADLs). An assistive system can help satisfy the demands of regular ADLs for people with disabilities. With an increasing shortage of caregivers and a growing number of individuals with impairments and the elderly, assistive robots can help meet future healthcare demands. One of the critical aspects of designing these assistive devices is to improve functional independence while providing an excellent human–machine interface. People with limited upper limb function due to stroke, spinal cord injury, cerebral palsy, amyotrophic lateral sclerosis, and other conditions find the controls of assistive devices such as power wheelchairs difficult to use. Thus, the objective of this research was to design a multimodal control method for robotic self-assistance that could assist individuals with disabilities in performing self-care tasks on a daily basis. In this research, a control framework for two interchangeable operating modes with a finger joystick and a chin joystick is developed where joysticks seamlessly control a wheelchair and a wheelchair-mounted robotic arm. Custom circuitry was developed to complete the control architecture. A user study was conducted to test the robotic system. Ten healthy individuals agreed to perform three tasks using both (chin and finger) joysticks for a total of six tasks with 10 repetitions each. The control method has been tested rigorously, maneuvering the robot at different velocities and under varying payload (1–3.5 lb) conditions. The absolute position accuracy was experimentally found to be approximately 5 mm. The round-trip delay we observed between the commands while controlling the xArm was 4 ms. Tests performed showed that the proposed control system allowed individuals to perform some ADLs such as picking up and placing items with a completion time of less than 1 min for each task and 100% success.

Human–machine interface-based wheelchair control using piezoelectric sensors based on face and tongue movements

“…For practical use in severe disabilities, we have mainly developed assistive, adaptive, and rehabilitative devices by employing biomedical signals and physical signs. Examples of commercial HMI systems to support people with movement disabilities in their daily activities include voice recognition [ 9 , 10 ], Sip and Puff switch using air pressure from lungs [ 11 ], Chin switch [ 12 ], and head pointer [ 13 , 14 , 15 ], which employ head–neck movements and eye trackers using cameras to detect pupil movement and direction [ 16 ]. However, an HMI system that covers all levels of disability still needs to be developed.…”