Resolving Observer Motion by Object Tracking

A fundanental sub-problem of the task of interpreting the sonar data gathered by a mobile subnersible robot vehicle is that of constructing and maintaining a stationary global reference frame in which interpretation can take place. Such a frane nay be established by tracking the movements of objects in the local environment of the robot and using the data so obtained to deduce the motion of the observer. In a marine context, where currents make it impossible to hold a vehicle stationary for any extended period, it is important to track the vehicle and object motion continuously in real time. In this paper the problem of establishing a stationary reference frame using noisy, specular, intermittent sonar data is discussed and a computational method for i t s solution is presented. The nethod has been tested by computer simulation in a two dimensional context and an evaluation of i t s performance is given. The solution presented is not restricted to marine vehicles but is espec ia l l y effective for land-based vehicles, where operating conditions are less severe.