Project Information

Summary

The fraction of population with physical disabilities has earned more relevance and has attracted the attention of international health care organizations, universities and companies interested in developing and adapting new products. The actual tendency reflects the demand of an increase on health and rehabilitation services, in a way that senior and handicapped individuals, who have been at the margin of society for a long period, might become more and more independent performing quotidian tasks.

Consequently, regardless the age, mobility is a fundamental characteristic for every human being. Children with disabilities are very often deprived of important opportunities and face serious disadvantages compared to other children. Adults who lose their independent means of locomotion become less self sufficient, raising a negative attitude towards themselves. The loss of mobility originates obstacles that reduce the personal and vocational objectives. Therefore is necessary to develop technologies that can aid this population group, in a way to assure the comfort and independence of the elderly and handicapped people. Wheelchairs are important locomotion devices for those individuals.

There is a growing demand for safer and more comfortable wheelchairs, and therefore, a new Intelligent Wheelchair (IW) concept was introduced. However, most of the Intelligent Wheelchairs developed by distinct research laboratories, have hardware and software architectures too specific for the wheelchair model used/project developed and are typically very difficult to configure in order for the user to start using them.

The importance and concern given to the autonomy and independence of elderly people and patients suffering from some kind of disability has been growing significantly in the last few decades. Intelligent wheelchairs (IW) are technologies that can increase the autonomy and independence of this kind of population and are nowadays a very active research area. However, the adaptations to users’ specificities and experiments with real users are topics that lack for deep studies.

This work developed a generic and modular framework for the development of IW based on multi-agent systems (MAS). The MAS developed is based on the concepts of micro and macro agent. Each IW represents a macro agent in the global MAS, interacting with the other agents (other IW, intelligent doors, among others). However, internally, each IW is a MAS composed by several modules distributed in autonomous agents - micro-agents. The communication platform for mobile robots developed is based on the concept of cooperation maintaining the global organization of the system and enforcing the concern with critical safety of the global system.

The intelligent wheelchair, developed in the context of the IntellWheels project, is controlled at a high-level by a flexible multimodal interface, using voice commands, facial expressions, head movements and joystick as its main inputs. This enables a flexible interface enabling a more natural interaction between the user and the IW through the use of input sequences composed by several inputs that trigger distinct actions of the device.

We also developed an automatic patient classification system enabling the automatic configuration, to the user characteristics, of the intelligent wheelchair. Therefore it was developed a methodology enabling to create a user model. The research was based on the development of a data gathering system, enabling the collection and storage of data from voice commands, facial expressions, head and body movements from several patients with distinct disabilities such as Cerebral Palsy.

The wheelchair can be used in different situations in real and simulated environments and a serious game was developed where different tasks should be performed by users. Data was analysed using knowledge discovery methods in order to create an automatic patient classification system. Based on the classification system, a methodology was developed enabling to select the best wheelchair interface and command language for each patient.

Experiments were conducted, using a large set of patients suffering from severe physical constraints in close collaboration with Escola Superior de Tecnologia de Saúde do Porto and Associação do Porto de Paralisia Cerebral. The experiments using the intelligent wheelchair were followed by user questionnaires.

The results were statistically analyzed in order to prove the effectiveness and usability of the adaptation of the Intelligent Wheelchair multimodal interface to the user characteristics. The results obtained in a simulated environment showed a 67 score on the system usability scale based in the opinion of a sample of cerebral palsy patients with the most severe cases IV and V of the Gross Motor Function Scale. It was also statistically demonstrated that the data analysis system advised the use of an adapted interface with higher evaluation than the one suggested by the occupational therapists, showing the usefulness of defining a command language adapted to each user.

Experiments conducted with distinct control modes revealed the users' preference for a shared control with an aid level taking into account the level of constraint of the patient. In conclusion, this work demonstrates that it is possible to adapt an intelligent wheelchair to the user with clear usability and safety benefits.

Objectives

IntellWheels Project main objective was the development of a wheelchair prototype that may help people with Quadriplegia and severe Cerebral Palsy to live a more normal life. The project specific objectives are:

  • To develop facial expression recognition methodologies appropriate for detecting expressions of cerebral palsy and quadriplegic individuals.

  • To build an intelligent wheelchair prototype including advanced sensorial capabilities, the use of computer vision as an aid for navigation.

  • To include intelligent planning and navigation capabilities in the wheelchair prototype.

  • To develop a high-level wheelchair command language enabling to fully control the device using simple commands and to interact with other physical devices such as other intelligent wheelchairs in the environment.

  • To project and implement a wheelchair high-level multimodal control application with interface based on facial expressions and other inputs. The interface will enable the user to fully configure the system selecting which facial expressions or inputs to use for triggering each command.

  • To develop a patient classification module based on machine learning and data mining techniques, enabling to automatically detect and classify the type of patient using the wheelchair.

  • To develop a simulator that enables IW simulation and mixed reality experiments;

  • To develop simple multi-robot coordination methodologies that will enable the IW to communicate with other IW and other devices and jointly solve specific planning problems arising in its environment.

  • Test and validate the system with a set of subjects suffering from severe cerebral palsy, quadriplegia and other physical disabilities.

The project final objective was concerned with creating the basis for the development of a new concept of configurable wheelchair that will enable the user to fully control the way the chair is controlled.