Publicação
Contributions for adapting electric wheelchairs to people with reduced handling capabilities
| Resumo: | The use of a commercial powered wheelchair during daily activities by people with motor limitations provides autonomy, increasing quality of life. For people with more severe restrictions, specifically reduced handling, the use of the powered wheelchair may be compromised as they cannot control the joystick with the required ability. This work is intended to give these people the ability to drive comfortably and safely their electric wheelchair. This proposal for adapting a commercial powered wheelchair has essentially two main objectives. The first one is to allow people who cannot drive a conventional powered wheelchair because they have muscle weakness in their upper limbs to do so. So, and to facilitate the wheelchair navigation, three modes of operation are proposed. Therefore, the adapted wheelchair controller must able translate the new impulses provided by the joystick, depending of the user abilities, in direction commands. Driving the powered wheelchair continuously can quickly become a stressor. Thus, the second goal is to ensure that during the powered wheelchair driving, its user always remains in a calm state, avoiding harming their well-being and their health. It is thus proposed to insert a system for monitoring the emotional state of the wheelchair user. The primary purpose of this system is to alert its user when he reached critical emotional levels, which makes it impossible to continue driving safely. In this situation the wheelchair controller becomes responsible for the wheelchair command. The implementation of the stress detection system was based in the galvanic skin response and heart rate sensors. With the acquired information, it was defined the user’s emotional profile and evaluated several algorithms classification to predict new emotional states. This system was tested by two users groups. The first group consisted of drivers driving cars and the second group consisted of tetraplegic people driving their powered wheelchair. For people without disabilities, it was confirmed through biometric signals that each person reacts differently to the same situations. This difference is most striking in the response of skin conductivity. For tetraplegic people, we concluded and confirmed, that the sympathetic nervous system response cannot be measured by galvanic skin response. Several classification algorithms were trained for stress detection in realtime. For each user, the best classifying model was found. We conclude that the best models vary from person to person and also depend on the number and type of predictors used. This work results from the collaboration of the University of Aveiro and the Centro de Reabilitação e Medicina da Região Centro - Rovisco Pais. |
|---|---|
| Autores principais: | Urbano, Maria Margarida Carreira Pires |
| Assunto: | Powered wheelchairs Joystick Handling Stress Galvanic skin response Heart rate Emotional profile Classification algorithms |
| Ano: | 2020 |
| País: | Portugal |
| Tipo de documento: | tese de doutoramento |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade de Aveiro |
| Idioma: | inglês |
| Origem: | RIA - Repositório Institucional da Universidade de Aveiro |
| Resumo: | The use of a commercial powered wheelchair during daily activities by people with motor limitations provides autonomy, increasing quality of life. For people with more severe restrictions, specifically reduced handling, the use of the powered wheelchair may be compromised as they cannot control the joystick with the required ability. This work is intended to give these people the ability to drive comfortably and safely their electric wheelchair. This proposal for adapting a commercial powered wheelchair has essentially two main objectives. The first one is to allow people who cannot drive a conventional powered wheelchair because they have muscle weakness in their upper limbs to do so. So, and to facilitate the wheelchair navigation, three modes of operation are proposed. Therefore, the adapted wheelchair controller must able translate the new impulses provided by the joystick, depending of the user abilities, in direction commands. Driving the powered wheelchair continuously can quickly become a stressor. Thus, the second goal is to ensure that during the powered wheelchair driving, its user always remains in a calm state, avoiding harming their well-being and their health. It is thus proposed to insert a system for monitoring the emotional state of the wheelchair user. The primary purpose of this system is to alert its user when he reached critical emotional levels, which makes it impossible to continue driving safely. In this situation the wheelchair controller becomes responsible for the wheelchair command. The implementation of the stress detection system was based in the galvanic skin response and heart rate sensors. With the acquired information, it was defined the user’s emotional profile and evaluated several algorithms classification to predict new emotional states. This system was tested by two users groups. The first group consisted of drivers driving cars and the second group consisted of tetraplegic people driving their powered wheelchair. For people without disabilities, it was confirmed through biometric signals that each person reacts differently to the same situations. This difference is most striking in the response of skin conductivity. For tetraplegic people, we concluded and confirmed, that the sympathetic nervous system response cannot be measured by galvanic skin response. Several classification algorithms were trained for stress detection in realtime. For each user, the best classifying model was found. We conclude that the best models vary from person to person and also depend on the number and type of predictors used. This work results from the collaboration of the University of Aveiro and the Centro de Reabilitação e Medicina da Região Centro - Rovisco Pais. |
|---|