A recent paper by members of the DCIST alliance addresses the problem of maintaining resource availability in a networked multi-robot team performing distributed tracking of an unknown number of targets. Robots receive and process sensor measurements locally and exchange information to cooperatively track a set of targets using a distributed Probability Hypothesis Density (PHD) filter. Each robot’s sensor measurement noise covariance matrix is used to quantify its sensing quality. If a robot’s sensor quality degrades, the team’s communication network is minimally reconfigured such that the robot with the faulty sensor may take better advantage of information from other robots to lessen the impact on overall team tracking performance. without enforcing a large change in the number of active communication links. Two PHD fusion methods are studied and four different Mixed Integer Semi-Definite Programming (MISDP) formulations (two formulations for each fusion method) are proposed to solve the problem.
Figure: Resilient target tracking. (Left) One of the tracker robots in the team experiences a sensor fault, thereby affecting the team’s tracking performance. (Right) The team has reconfigured to realize a new communication graph in 3D. Tracking performance has improved compared to what it was immediately after the robot’s sensor fault.
Points of Contact: Gaurav Sukhatme (PI), Ragesh Kumar Ramachandran
Citation: R. K. Ramachandran, N. Fronda and G. S. Sukhatme, “Resilience in Multirobot Multitarget Tracking With Unknown Number of Targets Through Reconfiguration,” in IEEE Transactions on Control of Network Systems, vol. 8, no. 2, pp. 609-620, June 2021, doi: 10.1109/TCNS.2021.3059794.