A Motion Planning Strategy for Rapidly Finding an Object with a Mobile Manipulator in 3-D Environments

In this paper, we address the problem of searching for an object in a 3-D environment. We consider a mobile manipulator with an "eye-in-hand" sensor moving in the 3-D environment. In particular; we consider a. static object whose location is modeled with a probability density function (pdf). We generate routes that minimize the expected value of the tune until the object is first seen when following the route. We use a sample-based convex cover to estimate the size and shape of visibility regions in 3-D. The resulting convex regions are exploited to generate trajectories that compromise between moving the manipulator base and moving the robotic arm.