CompanionAble Project


What is the Companionable Project? Ponder this: is a companion robot a friend, caregiver or a butler? In one study about human reactions to human-robot interactions, results showed that a large proportion of participants were in favor of robot companions as being assistants or servants, and only a few wanted robot companions as friends. 


The Aliz-E Robot Companion

The Aimec Robot Companion

Cognitive Robots Companions

Also, humans preferred robots with the ability to complete household tasks over the ability to carry out child and animal care tasks. Lastly, humans preferred companion robots that have the capacity for human-like communication. Human-like behavior and appearance were apparently not so important in this CompanionAble Project.

It turns out though, that humanlike behavior, at least on some basic cognitive level is important, especially to be an effective robot companion. In fact, companion or carebots are rapidly being associated with cognitive robots, or robots that have some degree of human intelligence. Dr Raja Chatila, a research director, says that there are four capacities required by a cognitive robot companion: perception and cognition of environment; learning by observation; decision making; communication and interaction with humans.

"Getting a robot to move around a human, without hurting them, and while making them feel comfortable, is a vital task," says Dr Chatila. Equally, for a cognitive robot to work, it must be able to pick up on subtle cues. If a human leans forward to get up, for example, the robot must understand the purpose of that movement. Much of human communication relies on non-verbal cues, and cognitive machines need to be able to understand these if they are to be useful, rather than irritating.

Even in verbal communication there are several habits that robots need to acquire. Most communicative habits of humans are so second nature to us that we never think of them. Dr. Chatila of the CompanionAble Project says that turn taking during conversations, for example is one thing they need to make robots understand. Humans take turns to talk; we need to find a way to make robots do the same," says Dr Chatila.

In Europe, some of the biggest advances in cognitive robotics are being made. One such project is the CompanionAble robot, or Integrated Cognitive Assistive & Domotic Companion Robotic Systems for Ability & Security, in which 18 European partners from seven countries are taking part. Lead by the University of Reading in the United Kingdom, the aim of the Companionable project is to “contribute to the enhancement of the quality of life of elderly and disabled persons using robotized solutions designed to operate in intelligent homes.” Specifically, CompanionAble addresses the issues of “social inclusion and homecare of persons suffering from chronic cognitive disabilities,” a condition prevalent among the increasing European older population.

As such, there is a great need to find solutions that will help the elderly in Europe live at home (semi)-independently for as long as possible. It is widely known that without the proper cognitive stimulation support, dementia and depression sufferers, especially the elderly, can deteriorate rapidly, demanding more from their carers. One survey undertaken by the Centro de Investigaciones Científicas (CIS) in Spain revealed that more than 80% of elderly persons over 65 prefer to stay in their own homes, even when they are in need of help.

The initiative, part of the European Union 7th Framework Program, hopes to address this by combining the use of robots in intelligent domestic environments, with the goal of creating a companion that assists people in their homes and helps them to be independent. Its main lines of research involve the development of technologies to build a system conscious of its surroundings, using a network of sensors and communications. From this they hope to produce mobile robots aware of the presence of persons within an intelligent and domotic (home-automated) household, and that are able to interact with them in a natural and intuitive form. The goal is to generate a simple and intuitive relationship between person and machine. For this, one partner of the project, Tecnalia Health and Quality of Life Unit, is developing a ‘robotic body language’ and a ‘robotic facial expression’ in order to communicate, in a non-verbal manner, the internal state of the robot.

Concretely, these robots are assigned to manage the daily life of the elderly or disabled persons, generate content for their cognitive stimulation, give reminders to take prescribed medication and analyze data on the state of their health. The robot will have an avatar-like robot face, an easy-to-use, natural interface that will facilitate hassle-free human-robot interaction and will be programmed not to bump into anything, especially its patient.

The system is able to detect and monitor people using various types of sensors (vision, sound and measurements of distance) which are also used to recognize emotional states and identify differences between normal and exceptional behavior. The patient may communicate with the robot using either a tactile screen or through more natural means, such as dialogue, facial expression or body movements. To achieve this, the environment-aware system of the robot has to have recognition of verbal and non-verbal instructions, as well as processes of cognitive stimulation. The technologies of the CompanionAble include developing a system prototype that could record emotions using audiovisual means, an essential component for intelligent dialogue and cognitive stimulation between the robot and the patient.

Other partners of the CompanionAble project are also investigating techniques for easier interaction between the assisted person and the robot and between the assisted person and his or her external carers using remote control. These techniques involve adapting portable sensors to identify the state of health of the user; designing communications protocols; gathering data and methodologies for signal analysis; and the development of an effective interface for external carers and technical services’ providers.

One of the system’s most import features is the monitoring of sensors that record the user’s everyday routine (Life Style Monitoring or LSM). The LSM records the activities and routines of the user in his or her home, by reading the various domestic devices in the domotic household. The LSM learns and memorizes the habits of persons, using a passive monitoring of the domestic environment. The sensors then recognize situations which deviate from the normal routine of the user. Irregular habits can then be interpreted as symptoms of a health problem or crisis, in which case the machine can activate an alert directed at a carer or a family member. Aside from providing therapeutic treatment and care support and social and health services, the robot is also designed to support a videoconferencing feature for communication with family members or professional carers.

Definitely sounds like a friend, caregiver and butler all in one in this CompanionAble Project.