I. Introduction
Object recognition within environments with large and dynamic occlusions is a challenging task that can be tackled by either deploying an extensive and expensive sensor constellation or by actively moving a set of sensors within the environment in order to maximize the observable surface area of the target objects. This is a variant of the View Planning Problem (VPP) [1] , which has a wide range of applications within the active perception domain, such as the estimation of the next best view for 3D scanning [2] , object recognition with occlusions, exploration of unknown environments, deployment of sensor networks to monitor targets, among many others.