Abstract
A robot is a machine including a number of sensors and a number of actuators. A living creature has also a number of sensors and a number of actuators. They help to survive in different kinds of environments. There are many kinds of creatures with many kinds of sensors and actuators which can survive in many different kinds of environments. Many of the living creatures developed a neuron based structure: a brain. Some of them (as humans) developed a very complex structure which allows them not only to react to the things sensed from their sensory system, but also to learn new abilities, remember things, have emotions, reason, etc. The term mind is used to refer to such higher functions of the human brain. Although no body totally understands how the mind really works, many theories have been proposed about how it may work. One of them is presented in the book The Society Of Mind of Marvin Minsky, a mathematician who is considered one of the fathers of Artificial Intelligence. Antonio Damasio, Professor of Neurology, in his book Descartes’ Error coincides in many details with the view of Minsky about how mind may work.Minsky uses the following words to describe the essence of how mind may work:
“I’ll call “Society of Mind” this scheme in which each mind is made of many smaller processes. These we’ll call agents. Each mental agent by itself can only do some simple thing that needs no mind or thought at all. Yet when we join these agents in societies—in certain very special ways—this leads to true intelligence. ”
—Marvin Minsky, “The Society of Mind”, pp 17.
The Society Of Mind has served as source of inspiration to develop Societies of Agents which have been implemented in the robot MiReLa (a RWI, B21 robot) and in the lamp MiFaRe, (a lamp operated by a musical language). This societies are formed from agents which can do very simple things by themselves, the communication between them is also simple and limited, but all together make the machines able to achieve progressively more complex tasks.
For the interaction from people to the machines and from machines to the people, sound generation/recognition has been chosen as the main interaction way. It is a way we humans share with many living creatures to exchange information remotely. Humans developed spoken languages, which, coding/decoding sound according to complex phonetic and grammatical rules, makes possible a very rich and effective interaction. The problem is that nowadays there is nothing available to help decode reliably any of these spoken
languages artificially in no controlled environments. Nor is clear that such a complex coding/decoding system is necessary to support simple interaction with machines such as, for example, to switch the lights on and off. An artificial music language named MiReLa Music Language has been created which uses music to support people-machine interaction. The MiReLa Music Language can be played by many musical instruments, mobile phones and PDAs, and it can be whistled. It has been used to interact with the robot MiReLa and to operate a lamp and it has probed to be robust and reliable in noisy environments.
This thesis proposes Societies of Agents to be implemented in machines or sensor networks (such as those used in Ambient Intelligence) to make it possible to incrementally add new capabilities and suggests one way to build them, keeping the agents and the communication between them simple. For people-machine, machine-machine interaction, it proposes a Musical Language, which is easy to whistle.
| Date of Award | 2005 |
|---|---|
| Original language | English |
| Awarding Institution |
|