Objectives
The general objective of the project is to develop an open knowledge environment for self-configurable, low-cost and robust robot swarms usable in domestic and public area applications - cleaning, patrolling, semantic mapping, escorting and others.
The targeted solution integrates local knowledge bases of robots and global web services via ad hoc communication and decentralized autonomic service discovery and composition mechanism.
The use of local and global level mission related knowledge is pursued by creating on-site distributed data environment on RFID tags and on-board databases that is supported by on-demand wireless access to global web resources.
System is kept scalable and manageable by decentralization of control and manipulation to local level.
By the end of the project, a demonstration of the proposed approaches will be implemented in the swarm carrying out patrolling and cleaning tasks in an emulated hospital environment.
The proposed ROBOSWARM environment is targeted to allows relatively simple swarm robots to:
- communicate on a large scale to divide tasks for increasing the functionality of the swarm (scalability);
- learn and share the knowledge with other swarm members via the global and/or local knowledge bases,
- bring together the scattered knowledge of the swarm to assure self-adaptation in dynamic and partially defined situations;
- operate successfully with minimal sensing capabilities (cost efficiency).
Technical objectives
- definition of the swarm (knowledge) environment architecture and the underlying (possibly spatial) computation infrastructure
- definition of high level knowledge exchange format (language) that is universal, semantically machine understandable and suitable for vendor independent inter-robot communication;
- on-site truly distributed knowledge base on RFID tags;
- robotic web-services for global knowledge reuse and exchange between swarms;
- embedded and server-oriented self-learning and decision making solutions that assure the consistency of intra- and inter-swarm behaviour;
- validation of research results in a real environment engaging stakeholders into the process and developing recommendations for robot manufacturers.
