Intelligent Single Switch Wheelchair Navigation

We have developed an intelligent single switch scanning interface and wheelchair navigation assistance system, called ISSWN, to improve driving safety, comfort, and efficiency for individuals who rely on single switch scanning as a control method. ISSWN combines a standard powered wheelchair with a laser rangefinder, a single switch scanning interface and a computer. It provides the user with context sensitive and task specific scanning options that reduce driving effort, based on an interpretation of sensor data together with user input. Trials performed by 9 able-bodied participants showed that the system significantly improved driving safety and efficiency in a navigation task. Introduction Mobility limitations are not only strong predictors of difficulties with activities of daily living, but also act as a primary factor to decrease educational, social, and vocational opportunities, which can result in social isolation, anxiety, and depression [1]. Conventional manual and powered wheelchairs or scooters are good alternatives for many people with physical disabilities. However, some people who have severely impaired motor functions or have a combination of multiple disabilities have found it difficult or impossible to use traditional mobility devices independently [2, 3]. In order to accommodate this population, several research groups have designed smart wheelchairs, which combine a power wheelchair with a set of sensors and a computer [4]. However, while most efforts to develop smart wheelchair technology have focused on improving navigation such as assuring collision-free travel, aiding the performance of specific tasks, and autonomously transporting the user between locations [4], research focusing on input methods for smart wheelchairs has been less frequent. In addition to a joystick, several kinds of input method have been used to operate smart wheelchairs, including voice recognition [5], sight path tracking [6], machine vision [7, 8], and single switch scanning. Single switch scanning is one of the least efficient methods for people with severe motor impairments to operate power wheelchairs. Problems with single switch wheelchair navigation include: frequent stops to counteract drift and to avoid obstacles, increased driving time, and frustration and fatigue in challenging environments such as narrow hallways. Several approaches to improve performance with single switch scanning interface have been investigated, [9-14] but most of this research focused on automating scan rate adjustment for entering text into computers and augmentative and alternative communication devices, not on wheelchair operation. We have developed an intelligent single switch scanning interface and wheelchair navigation assistance system, called ISSWN, to improve driving safety, comfort, and efficiency for individuals who rely on single switch scanning as a control method. As shown in Figure 1, ISSWN combines a standard powered wheelchair with a laser rangefinder, a single switch scanning interface and a computer. The laser rangefinder is used to identify features in the environment, including walls, doors and obstacles. A single switch is used as the user input device and a small laptop computer as an interface and processor. Figure 1. Intelligent Single Switch Wheelchair Navigation An algorithm was developed to assist with single switch wheelchair navigation by sharing control between the wheelchair driver and the system. As shown in Figure 2, the system infers the current context based on user input representing the user’s intention and sensor data rendering environmental features to provide the user with a context sensitive scanning interface. Yanco & Gips compared navigation performance using a single switch scanning interface between a smart wheelchair, which provided a user with wall following and obstacle avoidance, and a traditional powered wheelchair [15]. In the study, it was demonstrated that a smart wheelchair significantly surpassed a general powered wheelchair in drive performance using single switch scanning. However, while the single switch scanning interface they used provided a static scanning interface, ISSWN provides the user with context sensitive options by dynamically adjusting its single switch scanning interface and its behavior based on the user's input and the wheelchair's surroundings. As shown in Figure 3 and Figure 4, the user interface provided by ISSWN has three different modes: manual mode, context sensitive mode, and autonomous mode. Figure 3. User Interfaces provided by ISSWN Figure 2. Intelligent Algorithm of ISSWN Figure 4. Activity Diagram of ISSWN In manual mode, the ISSWN makes the wheelchair stop when it detects objects or a drop-off within a set range and then switches to context sensitive mode. In context sensitive mode, ISSWN provides the user with task specific options as navigation assistance (Figure 3), including: • Obstacle avoidance – ISSWN makes the wheelchair turn left or right until it gets enough space to go forward or backward, depending on the user's input. Then it switches to autonomous mode. • Passing-through-doorway – ISSWN adjusts the minimum obstacle clearance used for obstacle avoidance when it detects two closely-spaced obstacles after activating the corresponding option in context sensitive mode. Then it switches to autonomous mode. Activity Diagram of ISSWN Context Sensitive Mode Manual Mode Autonomous Mode