The use of unmanned vehicles to perform tiring, hazardous, repetitive tasks, is becoming a reality out of the academy laboratories, getting more and more interest for several application fields from the industrial, to the civil, to the military contexts. In particular, these technologies appear quite promising when they employ several low-cost resource-constrained vehicles leveraging their coordination to perform complex tasks with efficiency, flexibility, and adaptation that are superior to those of a single agent (even if more instrumented). In this work, we study one of said applications, namely the visual tracking of an evader (target) by means of a fleet of autonomous aerial vehicles, with the specific aim of focusing on the target so as to perform an accurate position estimation while concurrently allowing a wide coverage over the monitored area so as to limit the probability of losing the target itself. These clearly conflicting objectives call for an optimization approach that is here developed: by considering both aforementioned aspects and the cooperative capabilities of the fleet, the designed algorithm allows controling in real time the single fields of view so as to counteract evasion maneuvers and maximize an overall performance index. The proposed strategy is discussed and finally assessed through the realistic Gazebo-ROS simulation framework.

When a robot has to interact with a person in a dynamic environment, it has to navigate to reach a close distance and to be in front of the person. This navigation has to be smooth and take care of the person's movements, the static obstacles and the motion of other people. In this paper, we present a new method to approach a person, that combines G 2 -Splines (G 2 S) paths with the Extended Social Force Model (ESFM) to allow the robot to move in dynamic environments avoiding static obstacles and other people. Moreover, we use the Bayesian human motion intentionally prediction (BMP) in combination with the Social Force Model (SFM) to be able to approach a moving person and also to avoid moving people in the environment. The method computes several paths using the G 2 S and taking into account the person's position and orientation. Then, the method selects the best path using several costs that consider distance, orientation, and interaction forces with static obstacles and moving people. Finally, the robot is controlled with the ESFM to follow the best path. The method was validated by a set of simulations and also by real-life experiments with a humanoid robot in a dynamic environment.