A cooperative trajectory generation framework for cluttered environments

This paper proposes a cooperative time-critical trajectory generation framework for multiple Unmanned Aerial Systems (UAS) operating in cluttered environments to achieve a common goal. The framework builds upon and combines several existing algorithms that are specifically designed to: i) reduce the data exchanged over a lossy wireless communication network, ii) allow efficient and safe operations in cluttered environments, and iii) account for the uncertainty associated with each obstacle in the scenario, depending on the source of information that provided the geometric description of the object. The first component of the framework is a smooth trajectory generation algorithm for multiple vehicles that leverages the geometric properties of a set of polynomial curves denoted as Pythagorean Hodographs, while extracting geometrically relevant information from obstacles in conflict. The second component of the framework is a time-critical coordination algorithm that modifies the speed profile of each UAS to ensure that the fleet is coordinated, and meets its temporal constraints in the presence of external disturbances. The path-following block ensures the vehicles converge to and follow their assigned paths, while the Control Augmentation System (CAS) translates the higher level commands produced by the time-critical coordination and path-following algorithms into actuator commands. The last component of the framework is an online monitoring tool that checks whether the temporal and spatial constraints of the fleet are still feasible.