Collision Avoidance
Some of the next generation massive spectroscopic survey projects, such as DESI and PFS, plan to use thousands of fiber positioner robots packed at a focal plane to quickly move in parallel the fiber-ends from the previous to the next target points. The most direct trajectories are prone to collision that could damage the robots and impact the survey operation.
In collaboration with REACT group at EPFL we developed a motion planning framework based on a novel decentralized navigation function for collision-free coordination inspired from solutions provided for traffic coordination of driverless vehicles at intersections (For more information on this project please visit: Hierarchical Traffic Coordination of Automated Guided Vehicles).
The method takes into account the configuration of positioners as well as the actuator constraints. We provide details for the proof of convergence and collision avoidance. Decentralization results in linear complexity for the motion planning as well as dependency of motion duration with respect to the number of positioners. Therefore the coordination method is scalable for large-scale spectrograph robots. The short in-motion duration of positioner robots, will thus allow the time dedicated for observation to be maximized.