New research reports shows that important progresses for controlling electric-powered wheelchair have been made recently aiming people with severe disabilities. In fact, a significant amount of people affected by those physical disabilities still cannot take advantage of autonomous mobility, even electronic or automated ones. For those people, the use of proper biological signals to control the assisted environment may be the only existing solution. In such scenario, the act of commanding an electric-powered wheelchair without proper training may be a serious safety risk. To avoid this kind of dangerous situation and to permit users to make use of such technology, one viable solution would be to be trained by using virtual driving simulators. Nevertheless, when using biomedical signals as commands it is not possible to ensure a continuous and reliable control of the wheelchair, it is necessary to associate the control possibilities with autonomous features such as semiautomatic obstacle detection or contour. Thus, it is interesting to offer to new wheelchair users the possibility of using simulators to allow them to learn to drive at distance, making use of telematics techniques combined with mobile and wearable devices, and publishing their progress and worries in social networks, which is the objective of this work. This project joints complementary skills from researchers from the Federal University of Amazonas, Federal University of Espirito Santo, and Federal University of Uberlandia, with the collaboration of researches from the University of Lorraine in Metz-France.

In this paper, an assessment of a driving assistance by a deictic command for a smart wheelchair is proposed. This equipment enables the user to move with a series of indications on an interface displaying a view of the environment and bringing about automatic movement of the wheelchair. Two sets of tests were implemented to assess the advantages of this type of assistance compared with conventional wheelchair control. The first set evaluated the performance of the human-machine system that is based on a course time analysis, an observation of users' actions, and an estimation of driving comfort. The second test was implemented to assess the cognitive requirements of the driving task, specifically the attentional and executive processes required when driving in assisted mode. A dual-task method was used to achieve this. The results show that driving assistance brings about a decrease in physical load for the same level of comfort as manual driving, but requires an additional cognitive effort for the user, especially in terms of executive abilities.

The VAHM project constitutes a research and development project aimed at automating an electric wheelchair for the disabled. The aim consists in delegating the control of some low-level functionalities to the wheelchair during the motion in order to allow the person to be free of the driving constraints that may be painful for some handicapped persons. The authors describe the hardware and software structure chosen to perform the intelligent mobility functions in association with a disabled person.>