Artificial intelligence research has foundered on the issue of representation. When intelligence is approached in an incremental manner, with strict reliance on interfacing to the real world through perception and action, reliance on representation disappears. In this paper we outline our approach to incrementally building complete intelligent Creatures. The fundamental decomposition of the intelligent system is not into independent information processing units which must interface with each other via representations. Instead, the intelligent system is decomposed into independent and parallel activity (...) producers which all interface directly to the world through perception and action, rather than interface to each other particularly much. The notions of central and peripheral systems evaporateeverything is both central and peripheral. Based on these principles we have built a very successful series of mobile robots which operate without supervision as Creatures in standard office environments. (shrink)

Most animals have significant behavioral expertise built in without having to explicitly learn it all from scratch. This expertise is a product of evolution of the organism; it can be viewed as a very long term form of learning which provides a structured system within which individuals might learn more specialized skills or abilities. This paper suggests one possible mechanism for analagous robot evolution by describing a carefully designed series of networks, each one being a strict augmentation of the previous (...) one, which control a six legged walking machine capable of walking over rough terrain and following a person passively sensed in the infrared spectrum. As the completely decentralized networks are augmented. the robot’s performance and behavior repetoire demonstrably improve. The rationale for such demonstrations is that they may provide a hint as to the requirements for automatically building massive networks to carry out complex sensory-motor tasks. The experiments with an actual robot ensure that an essence of reality is maintained and that no critical disabling problems have been ignored. (shrink)

In order to build autonomous robots that can carry out useful work in unstructured environments new approaches have been developed to building intelligent systems. The relationship to traditional academic robotics and traditional artificial intelligence is examined. In the new approaches a tight coupling of sensing to action produces architectures for intelligence that are networks of simple computational elements which are quite broad, but not very deep. Recent work within this approach has demonstrated the use of representations, expectations, plans, goals, and (...) learning, but without resorting to the traditional uses, of central, abstractly manipulable or symbolic representations. Perception within these systems is often an active process, and the dynamics of the interactions with the world are extremely important. The question of how to evaluate and compare the new to traditional work still provokes vigorous discussion. (shrink)

Both direct, and evolved, behavior-based approaches to mobile robots have yielded a number of interesting demonstrations of robots that navigate, map, plan and operate in the real world. The work can best be described as attempts to emulate insect level locomotion and navigation, with very little work on behavior-based non-trivial manipulation of the world. There have been some behavior-based attempts at exploring social interactions, but these too have been modeled after the sorts of social interactions we see in insects. But (...) thinking how to scale from all this insect level work to full human level intelligence and social interactions leads to a synthesis that is very di erent from that imagined in traditional Arti cial Intelligence and Cognitive Science. We report on work towards that goal. (shrink)

Artificial Intelligence as a discipline has gotten bogged down in subproblems of intelligence. These subproblems are the result of applying reductionist methods to the goal of creating a complete artificial thinking mind. In Brooks (1987) 1 have argued that these methods will lead us to solving irrelevant problems; interesting as intellectual puzzles, but useless in the long run for creating an artificial being.

The first part of this paper explores the general issues in using Artificial Life techniques to program actual mobile robots. In particular it explores the difficulties inherent in transferring programs evolved in a simulated environment to run on an actual robot. It examines the dual evolution of organism morphology and nervous systems in biology. It proposes techniques to capture some of the search space pruning that dual evolution offers in the domain of robot programming. It explores the relationship between robot (...) morphology and program structure, and techniques for capturing regularities across this mapping. (shrink)

Behavior based systems require an orthogonal view of integration issues. In this paper we highlight those issues, discuss what is easy, what is hard, and where the research frontiers lie.

We describe an algorithm which allows a behavior-based robot to learn on the basis of positive and negative feedback when to activate its behaviors. In accordance with the philosophy of behavior-based robots, the algorithm is completely distributed: each of the behaviors independently tries to find out (i) whether it is relevant (ie. whether it is at all correlated to positive feedback) and (ii) what the conditions are under which it becomes reliable (i.e. the conditions under which i t maximizes the (...) probability of receiving positive feedback and minimizes the probability of receiving negative feedback). The algorithm has been tested successfully on an autonomous 6-legged robot which had to learn how to coordinate its legs so as to walk forward. (shrink)

At the MIT Arti cial Intelligence Laboratory we have been working on technologies for an Intelligent Room. Rather than pull people into the virtual world of the computer we are trying to pull the computer out into the real world of people. To do this we are combining robotics and vision technology with speech understanding systems, and agent based architectures to provide ready at hand computation and information services for people engaged in day to day activities, both on their own (...) and in conjunction with others. We have built a layered architecture where at the bottom level vision systems track people and identify their activities and gestures, and through word spotting decide whether people in the room are talking to each other or to the room itself. At the next level an agent architecture provides a uniform interface to such specially built systems, and to other o the shelf software, such as web browsers, etc. At the highest level we are able to build application systems that provide occupants of the room with specialized services; examples we have built include systems for command and control situations rooms and as a room for giving presentations. (shrink)

We have previously built a small IKg ([Angle 89] and [Brooks 89]) six legged walking robot named Genghis. It was remarkably successful as a testbed to develop walking and learning algorithms. It encouraged us to build a more fully engineered robot with higher performance. We are building two copies of the robot, both 1.6Kg in mass. Their generic name is Attila. Attila has 24 actuators and over 150 sensors, all connected via a local network (the I2C bus) to 11 onboard (...) computers. (shrink)

We present a novel methodology for building humanlike artificially intelligent systems. We take as a model the only existing systems which are universally accepted as intelligent: humans. We emphasize building intelligent systems which are not masters of a single domain, but, like humans, are adept at performing a variety of complex tasks in the real world. Using evidence from cognitive science and neuroscience, we suggest four alternative essences of intelligence to those held by classical AI. These are the parallel themes (...) of development, social interaction, embodiment, and integration. Following a methodology based on these themes, we have built a physical humanoid robot. In this paper we present our methodology and the insights it affords for facilitating learning, simplifying the computation underlying rich behavior, and building systems that can scale to more complex tasks in more challenging environments. (shrink)

Complex systems and complex missions take years of planning and force launches to become incredibly expensive. The longer the planning and the more expensive the mission, the more catastrophic if it fails. The solution has always been to plan better, add redundancy, test thoroughly and use high quality components. Based on our experience in building ground based mobile robots (legged and wheeled) we argue here for cheap, fast missions using large numbers of mass produced simple autonomous robots that are small (...) b y today's standards (1 to 2 Kg). We argue that the time between mission conception and implementation can be radically reduced, that launch mass can be slashed, that totally autonomous robots can be more reliable than ground controlled robots, and that large numbers of robots can change the tradeoff between reliability of individual components and overall mission success. Lastly, we suggest that within a few years it will be possible at modest cost to invade a planet with millions of tiny robots. (shrink)

Both direct, and evolved, behavior-based approaches to mobile robots have yielded a number of interesting demonstrations of robots that navigate, map, plan and operate in the real world. The work can best be described as attempts to emulate insect level locomotion and navigation, with very little work on behavior-based non-trivial manipulation of the world. There have been some behavior-based attempts at exploring social interactions, but these too have been modeled after the sorts of social interactions we see in insects. But (...) thinking how to scale from all this insect level work to full human level intelligence and social interactions leads to a synthesis that is very different from that imagined in traditional Artificial Intelligence and Cognitive Science. We report on work towards that goal. (shrink)

There are a number of reasons to be interested in building humanoid robots. They include (1) since almost all human artifacts have been designed to easy for humans to interact with, humanoid robots provide backward compatibility with the existing human constructed world, (2) humanoid robots provide a natural form for humans to operate through telepresence since they have the same kinematic design as humans themselves, (3) by building humanoid robots that model humans directly they will be a useful tool in (...) understanding how humans develop and operate as they provide a platform for experimenting with different hypotheses about humans and (4) humanoid robots, given su cient abilities, will present a natural interface to people and people will be able to use their instinctive and culturally developed subconscious techniques for communicating with other people to communicate with humanoid robots. In this paper we take reason (4) seriously, and examine some of the technologies that are necessary to make this hope a reality. (shrink)

Applications of learning to autonomous agents (simulated or real) have often been restricted to learning a mapping from perceived state of the world to the next action to take. Often this is couched in terms of learning from no previous knowledge. This general case for real autonomous robots is very difficult. In any case, when building a real robot there is usually a lot of a priori knowledge (e.g., from the engineering that went into its design) which doesn’t need to (...) be learned. We describe the behavior-based approach to autonomous robots, and then examine four classes of learning problems associated with such robots. (shrink)

We want to build tiny gnat-sized robots, a millimeter or two in diameter. They will be cheap, disposable, totally sefcontained autonomous agents able to do useful things in the world. This paper consists of two parts. The first describes why we want to build them. The second is a technical outline of how to go about it. Gnat robots are going to change the world.

This paper presents an autonomous vision-based obstacle avoidance system. The system consists of three independent vision modules for obstacle detection, each of which is computationally simple and uses a di erent criterion for detection purposes. These criteria are based on brightness gradients, RGB Red, Green, Blue color, and HSV Hue, Saturation, Value color, respectively. Selection of which modules are used to command the robot proceeds exclusively from the outputs of the modules themselves. The system is implemented on a small monocular (...) mobile robot and uses very low resolution images. It has been tested for over 200 hours in diverse environments. Keywords: Vision-based navigation, space exploration, modular design, reactive control, unstructured terrain. (shrink)