Motivation Scenario

Natural collective systems demonstrate that many individuals cooperate when it is profitable to every one of them to do so. Examples are cooperative hunting of predatory animals, group-based foraging of mammals or nest building of social insects. In these and many other examples, animals get together when doing so provides better chances for foraging, defence, or generally surviving in their environment. Members of these groups can be weak with limited sensors/actuator capabilities, however collectively they can build a strong group with very extended capabilities.
Lately, technical systems mimic natural collective systems for the improvement in functionality of artificial swarm agents. Collective, networked or swarm robotics are scientific domains, dealing with cooperation between robots. Research in collective robotics is mostly concentrated on stand-alone autonomous robots. Cooperation and competition among stand-alone robots increases their collective fitness. However, natural swarm agents can build principally new kinds of collective systems. For example, fungi dictyostelium discoideum can aggregate into a multi-cellular symbiotic organism and perform such activities that cannot be fulfilled alone or in a swarm-like way. The symbiotic organization emerges new functional capabilities that allow swarm agents to achieve better fitness in the environment. When the need for aggregation is over, the symbiotic organism can disaggregate and exists one again as a stand-alone agents.
Swarm robots can also build symbiotic life forms and therefore achieve better functional fitness. To demonstrate this idea, we consider a collective energy foraging in real swarm of microrobots Jasmine.

The robots, equipped with on-board recharging electronics, possess its own energy homeostasis. When swarm robots become "hungry", they can collectively look for energy resources (in this case the docking stations. The clever collective strategy can essentially improve the efficiency of energy foraging.
Now, the recharging station is separated from a working area by a small barrier, so that robots can never reach the energy source.

However robots can aggregate into more complex symbiotic organism (which has legs and can pass the barrier, they reach the docking station.

Thus, from aggregation into a symbiotic multi-robot organism new functionality emerges, which allows small robots to survive in the environment. Generally speaking, robots are heterogonous, e.g. with different sensors and actuators, so robots can help each another in a symbiotic life form. In this example we encounter new kinds of collective robotic systems:
  • adaptable functionality: small and simple robots are capable for autonomous aggregation and disaggregation into/from large and complex organisms. Small robots profit from the aggregation into organism by sharing energy, computational and communication resources and by emerging new functionalities. For example, the prototype of symbiotic robots, aggregating on the surface, can change its own shape, and in this way achieve different locomotion strategies:

  • dependable evolve-ability: the organism is controlled through a bio-inspired genetic framework. When robots aggregate into an organism, they build distributed middleware computational systems. The core of this system is a robot genome. By means of recombination and genetic learning, the organism is capable of self-adaptation, self-changing of functional shapes, change of locomotion principles and, finally, for self-programming in new environments. In this way the organism can self-adapt and self-program for working in unpredictable situations without human reprogramming. In this way, organisms can self-evolve to fit in best possible way their environment.
  • artificial immunology, embryology and virtual robot sexuality: the robot organism has a similar “construction” to other multi-cellular organisms, including also mammal animals. There are great questions about functionality of regulating functions of complex multi-cellular organisms, their reproductive systems and internal homeostasis. Different “robo-cells” in robot organism can also compete with each other, creating “cancer cells”, can exchange genetic information and artificial hormones. The whole artificial organism definitely needs internal regulating mechanisms (as demonstrated in preliminary experiments with Jasmine V robots). The SYMBRION project provides a chance to explore the workings of artificial immunological, embryological and reproduction mechanisms;
  • working environment: small robots of the organism represent nodes of the sensor network. The organism can perform autonomous placement and maintenance of sensor networks in different indoor working environments. Such a sensor network is adjustable such that the robot volume (size of the organism) can access different, even hard to reach, areas of monitoring. It has adaptable locomotion for movement in different types of terrain. It allows adapting to environment without human reprogramming, it is capable of autonomous energy supply and is it extremely reliable (even in non-hazardous environments). It can also cooperate with other robotic systems to exchange sensor data, world models and other sensor information and actuate cooperatively.

Created by admin. Last Modification: Friday 30 of November, 2007 13:30:11 CET by admin.