COGNIRON Cognitive Robot

COGNIRON Cognitive Robot Companions

COGNIRON Cognitive Robot Companion

The bar for science and technology keeps getting raised higher and higher. Apparently robots that know your voice and follow you around is not enough, for now scientists are working on plans to create the true robot companion: one who can serve as a personalized, life-long companion for the unique individual. In other words, robots who can actually adapt to the specific social, emotional and cognitive needs of the people they are living with. Sounds like a tough job. Husbands and wives have been trying to achieve this same exact thing for years.

COGNIRON Cognitive Robot Companion

Individualized robots are necessary due to the uniqueness of each human person: people have individual needs, likes and dislikes, preferences and personalities that a companion would have to adapt to, and therefore it is argued that one and the same robot will not fit all people. And thus the emergence of projects like COGNIRON, which is made up of European scientists and engineers that hope to make the leap from building single function “dumb” machines to adaptive learning machines: Machines that could really learn, robots that are not only ready-made devices but artificial creatures, which can improve their capabilities in a continuous process of acquiring new knowledge and skills

COGNIRON Cognitive Robot Companion

In the European Commission’s 'Beyond Robotics' work program, COGNIRON is said to be studying the development of cognitive robots whose “purpose in life” would be to serve humans as assistants or “companions.” Such robots would be able to learn new skills and tasks in an active open-ended way and to grow in constant interaction and co-operation with humans. The concept is classic science fiction fodder, but following the work of the COGNIRON cognitive robot companion project, which has been receiving funding from the IST’s Future and Emerging Technologies initiative since January 2004, this may one day become science fact.

So what can this COGNIRON cognitive robot companion do exactly?

Dr Raja Chatila, research director at the Systems Architecture and Analysis Laboratory of the French Centre National de la Recherche Scientifique (LAAS-CNRS), and COGNIRON project coordinator, says that the example often used for a cognitive robot is a robot that’s able to “fulfill your needs, like passing you a drink or helping in everyday tasks.” Of course once intelligent behavior is fully realized, there may be no limits as to the beneficial tasks a cognitive robot could do.

As of now, what it should be able to do is to act as a mechatronic device, endowed with mobility and manipulation capacities in human environments. But because the robot is meant for permanent interactions with humans in ever-changing contexts and situations, a cognitive companion robot simply cannot be a ready-made, “one-size-fits-all” device. This continued interaction with humans will always demand the execution of tasks that will continually change, and thus a robot companion must be prepared to deal with this kind of dynamism. The robot will then have to exhibit cognitive capacities for adapting its behavior in relation to what the situations and the specific task calls for.

To do this, a cognitive robot will have to understand its spatial surroundings, to make decisions according to its own evaluation of its situation, and to interact closely with humans both physically and through dialogue. These capacities cannot be pre-programmed, but will have to be learned. And so open-ended learning processes will have to be at the core of the robot’s every function, allowing the robot companion to successfully adapt to its human partner and learn new skills that will correctly match the humans' needs.

Dr Chatila summarizes the purpose of the project’s key research themes, “Research breaks down into four capacities required by a cognitive robot companion: perception and cognition of environment; learning by observation; decision making; communication and interaction with humans.” He says that decision-making is a fundamental capability of a cognitive robot. It is important for a robot’s autonomous deliberation, task execution, or for human-robot collaborative problem solving. It also integrates the three other capacities: interaction, learning and understanding the environment. He says that “getting a robot to move around a human, without hurting them, and while making them feel comfortable, is a vital task.”

It also means that a cognitive robot must be able to understand subtle cues and the purpose of a human’s movements. Since much of human communication is non-verbal, it is vital for cognitive machines to pick up cues that are important and what aren’t. To provide solutions to these problems, the researchers took inspiration from “natural cognition as it occurs in humans,” which is one reason why a cognitive robot companion needs to be able to learn.

As of now, the project has reportedly made enormous progress and the team feels confident they will meet their criteria for success: three “Key Experiments” are being implemented on real robots for the integration, demonstration and validation of the research results. One experiment will feature a robot building a model of its environment in the course of a home tour, another one will feature a curious and proactive robot that will be able to deduce that a human needs something to be done, and the third one will demonstrate a robot’s ability to learn by imitation and repetition.

The Cogniron project has already partially implemented all three experiments, but there is a long way to go before a fully functional Cognitive Robot Companion may be realized and potentially commercialized. Looks like we’ll have to make do with the people we live with for a little longer.