HOME JSK? Members Research Papers Platform Links

Go to Japanese page.

What is Remote Brained Robotics?

Robots of the future will act in the real world. To realize robots with the common sense to do this will, it is generally believed, require massively parallel processing. This is a problem for those of us who want to do experiments with robots in the real world today --- it is hard to build active, limber, situated robots when they have to carry along heavy brains. Our answer is ``remote-brained robots''.

A remote-brained robot does not bring its own brain within the body. It leaves the brain in the mother environment, by which we mean the environment in which the brain's software is developed, and talks with it by wireless links. In this 'remote-brained approach', we can design and implement a robot with both a large-scale powerful brain and a light-weight limber body. This opens up the possibility of working with intelligent behaviors of a multi-limbed robot such as a whole-bodied humanoid as yet not much studied.

In this framework, a body is defined as the real interface for a brain to the outer world. The channel from the body to the brain defines all the input to the brain. The brain is defined as a program which receives the input containing all the sensor information and sends the output for actuation to the body. The research task on intelligent behaviors with a remote-brained robot is to design and implement a brain program based on the defined interface between the brain and the body.

To put the brain inside the mother environment provides us with convenient and important advantages. It allows us to share the development environment for a brain and observe the internal states of the brain during experiments. The brain software is developed in the mother environment which is inherited over generations. It can benefit directly from the mother's `evolution', meaning that the software gains power easily when the mother is upgraded to a more powerful computer.

The remote-brained approach was motivated by the question of whether the inference and representation techniques of AI are useful for robots in the real world. The answer will depend on how AI is interfaced to the real world. The remote brained approach allows us to tie AI directly to the world, enabling the verification of high-level AI techniques which could previously only be used in simulation. For robot research, this approach opens the way to the use of powerful parallel computers. For AI, this approach allows experiments with real agents, an essential step to the application of AI in the real world.

There has been a missing link in research, between `AI which couldn't survive if embodied in the real world' and `robots with feeble intelligence'. Our approach thourgh building remote-brained robots aims to open the way for engineering advances which will bridge the gap.

    Remote-Brained Robotics: Interfacing AI with Real World Behaviors.
    In Proceedings of 1993 International Symposium on Robotics Research, 1993.

    M. Inaba, S. Kagami, F. Kanehiro, K. Takeda, T. Oka and H. Inoue
    Vision-Based Adaptive and Interactive Behaviors in Mechanical Animals using the Remote-Brained Approach
    Robotics and Autonomous Systems,Vol.17,No.1-2,pp.35-52,1996

    M. Inaba
    Remote-Brained Robots
    Proceedings of Fifteenth International Joint Conference on Artificial Intelligence (IJCAI-97),pp. 1593-1606, 1997