Project summary

The main scientific objective of the Swarm-bots project is to study a novel approach to the design and implementation of self-organising and self-assembling artefacts. This novel approach finds its theoretical roots in recent studies in swarm intelligence, that is, in studies of the self-organising and self-assembling capabilities shown by social insects and other animal societies.

The main tangible objective of the project is the demonstration of the approach by means of the construction of at least one of such artefact. We intend to construct a swarm-bot. That is, an artefact composed of a number of simpler, insect-like, robots(s-bots), built out of relatively cheap components, capable of self-assembling and self-organising to adapt to its environment.

The work will see the co-ordinated interplay of four main activities:

1. the design and implementation of the hardware,
2. the design and implementation of the simulator,
3. the design and implementation of the control system of the swarm-bot, and
4. the integration, testing and evaluation of the outcomes of these activities.

The general guidelines for designing the hardware, the simulator and the control mechanisms are jointly defined at the project start by the project participants, in order to ensure consistency across all levels, as decisions at one level impose constrains on what can be done at another level. The guidelines specify the appropriate hardware for s-bots, the characteristic required to the simulator (what the simulator should simulate) and the control mechanisms (selecting the appropriate space of distributed, swarm intelligence-based control mechanisms and the appropriate learning and evolutionary algorithms). Based on such specifications, prototypes will be developed, evaluated and documented for all three components: s-bots (hardware), simulation (software), and swarm-intelligence-based control mechanisms (software). The control mechanisms will be tested on both the simulator and the hardware implementations. Final versions of all the three components will be developed and fine-tuned using input and feedback from the other two components. Integrated prototype and demonstrator will be then developed. The main deliverable will be a set of hardware s-bots that can self assemble into a shape-changing swarm-bot to accomplish a small number of tasks. Tasks considered will be dynamic shape formation and shape changing and navigation on rough terrain. In both cases, will be considered situations in which a single s-bot cannot accomplish the task and the cooperative effort performed by the s-bots aggregated in a swarm-bot is necessary.

The expected result is a novel swarm intelligence-based method for the design and the implementation of self assembling and self-organising artefacts. Milestones will be prototypes of hardware, simulator, and control of swarm-bots.

Project objectives

This project aims to contribute to the economic development of the Community by providing a new approach to the design, construction and control of robotic systems.

The objective of the SWARM-BOTS project is to study a novel approach to the design, hardware implementation, test and use of self-assembling, self-organising, metamorphic robotic systems called swarm-bots. This novel approach finds its theoretical roots in recent studies in swarm intelligence, that is, in studies of the self-organising and self-assembling capabilities shown by social insects and other animal societies.

By the end of the project a demonstration of the proposed approach will be made by means of the physical construction of at least one swarm-bot, that is, a self-assembling and self-organising robot composed of a number (30-35) of smaller devices, called s-bots.

Because currently no such artifacts exists, a first step in measuring success is the realisation of a working swarm-bot. In particular, we intend to test the feasibility of the integration of swarm intelligence, reinforcement learning and evolutionary computation paradigms for the implementation of self-assembling and self-organising metamorphic robots by constructing a swarm-bot prototype. A major measurable goal will therefore be the (hardware) construction of such a prototype. The project will be considered successful if a working prototype capable of on-line self-organisation can be demonstrated at the end of the project. The working prototype will be able to achieve the following three sets of objectives:

Dynamic shape formation/change:

our swarm-bot, composed of at least twenty s-bots randomly distributed on the floor, will be able to self-assemble into a number of different planar and 3D geometric configurations, for example like those found in ant colonies and in patterns of differential adhesion by developing cells. These configurations will be closed shapes with internal structure, such as

1. centre/periphery figures (for example, all s-bots with a given set of sensors will stay on the outer perimeter whereas all other s-bots will remain inside);
2. checker-board;
3. split (each half of the assembly will contain s-bots with similar characteristics);

In addition, we shall test transitions between shapes. A long-term goal, but not necessary for the success of this project, is to achieve emergent expulsion of "dead bodies", that is of s-bots that are no longer functioning.

Navigation on rough terrain:

our swarm-bot, composed of at least twenty s-bots, will be capable of autonomously moving across the terrain guided by sensory information gathered by individual s-bots. In particular, we intend to achieve the following objectives:

1. light following while maintaining the original shape (for example, one of those described above);
2. light following through narrow passages and tunnels that require dynamic reconfiguration of the swarm-bot;
3. passing over a hole or through a steep concave region that could not be passed by a single s-bot;
4. moving from point A to B (for example, on a shortest possible trajectory) on rough terrain.

This latter point might include climbing/descending stairs, although not necessary for the success of the project. Some experiments (point 2 and 3) will be based on 3D structures.

Scaling up:

one of the above discussed experiments in both categories will be repeated with increasing number of s-bots (up to 30-35) to study the scaling behaviour of the swarm-bot robotic system.

The attainment of these objectives will be assessed through demonstrations and measurements. Measurements of goal attainment will be done using the following metrics:

1. the precision with respect to the desired shape (i.e., if we want the swarm-bot to acquire the shape of a square, how close to a square is the obtained shape),
2. the time necessary to form a shape and/or move the whole assembly from point A to B (ideally, this figure should be proportional -or better- to the number of s-bots),
3. the robustness to the addition and/or removal of s-bots, and
4. swarm-bot entropy as a measure of degree of organisation.

Project results

The results achieved during the course of the project are detailed in the Control Section. Here we present the main scenario we tackled.

An application in which self-organisation and self-assembling can be useful is the classical Search & Rescue in complex environments. A swarm of up to 35 s-bots must transport a heavy object from its initial to a goal location. On the right side can be seen the yellow goal location; on the left side the grey object to be transported, surrounded by s-bots. There are several possible paths from the initial to the goal location and these paths may have different lengths and may require avoiding obstacles and holes. The weight of the object is such that its transportation requires the coordinate effort of at least n s-bots, where n>1 is a parameter. The overall scenario can be split into two main tasks: finding an object or a goal location, and performing cooperative transport. This is detailed in the following.

	The s-bots are positioned in an arena and they have to locate and connect to the object to be transported. Other s-bots have to locate the goal and build a chain connecting it to the object's initial location.
At this phase, the s-bots after having located the goal and the optimal path to it, have formed a chain connecting the object to it. The chain members are the coloured s-bots. The rest of the s-bots are connected to the object and the transportation phase is ready to start.
	The object is been transported by the cooperative effort of the s-bots along the chain that has been established.
The task is nearly concluded as the object approaches the goal location.
Scenario 3D Overview Movie (MOV 11.5 MB) - This is an overview of the scenario in simulation
Scenario solved by 12 real s-bots (MPEG 42 MB)