The development and usage of UAVs quickly increased in the last decades, mainly for military purposes. Nowadays, this type of technology is also used in non-military contexts, for instance for environment protection, rescue services, fire fighters and police officers assistance, environmental scientific studies, etc. Although the technology for operating a single UAV is now mature, efforts are still necessary for using UAVs in fleets (or swarms). ASIMUT relies on the collaboration among UAVs evolving at different altitudes: fixed-wing UAVs covering the area from a high altitude with coarse grain acquisition capabilities and low-altitude multi-rotor UAVs in charge of precise measurements.
The goal of this project is to address scenarios where a mission has to be achieved based on decisions made from data collected from different sensors in a number of UAVs that constitute swarms. High-altitude and low-altitude linebreak swarms are tasked based on the fusion of information coming from multiple sources. From an early warning process where data are provided by the UAV swarm payloads, the target localization, monitoring and classification processes are achieved using advanced techniques from the domains of artificial intelligence, machine learning, and statistics.
The objectives of the ASIMUT project are to design, implement and validate algorithms that will allow the efficient usage of autonomous swarms of UAVs for surveillance missions. To support situation management, the ASIMUT project focuses on:
Providing automated assistance to human operators with regard to high level data fusion tasks. By doing so the project can significantly improve their capabilities for making efficient and effective decisions;
Developing methods and algorithms to improve moving target detection in a video flow;
Handling several swarms of UAVs including their communications, networking and positioning. This thus motivates the development of multilevel cooperation algorithms, which is an area that has not been widely explored, especially when autonomy is also a challenge. Thus, additional problematics appear, and it becomes necessary to focus on:
- Providing techniques to optimize communications within a swarm and between swarms (including multilevel swarms);
- Developing distributed and localized mobility management algorithms to cope with conflicting objectives, such as connectivity maintenance and geographical area coverage.