Towards Evolving Cooperative Mapping for Large-Scale UAV Teams

摘要

A team of UAVs has great potential to handle real-world challenges. Knowing the environment is essential to perform in an effective manner. However, in many situations, a map of the environment will not be available. Additionally, for autonomous systems, it is necessary to have approaches that require little energy, computing, power, weight and size. To address this, we propose a light-weight, evolving, and memory efficient cooperative approach for estimating the map of an environment with a team of UAVs. Additionally, we present proof-of-concept experiments with real-life flights, showing that we can estimate maps using an off-the-shelf web-camera.