Fusion of Inertial Sensing to Compensate for Partial Occlusions in Optical Tracking Systems

Abstract

Optical tracking is widely used for surgical Augmented Reality systems because it provides relatively high accuracy over a large workspace. But, it requires line-of-sight between the camera and the markers, which can be difficult to maintain. In contrast, inertial sensing does not require line-of-sight but is subject to drift, which causes large cumulative errors, especially for the measurement of position. This paper proposes a sensor fusion approach to handle cases where incomplete optical tracking information, such as just the 3D position of a single marker, is obtained. In this approach, when the optical tracker provides full 6D pose information, it is used to estimate the bias of the inertial sensors. Then, as long as the optical system can track the position of at least one marker, that 3D position can be combined with the orientation estimated from the inertial measurements to recover the full 6D pose information. Experiments are performed with a head-mounted display (HMD) that integrates an optical tracker and inertial measurement unit (IMU). The results show that with the sensor fusion approach we can still estimate the 6D pose of the head with respect to the reference frame, under partial occlusion conditions. The results generalize to a conventional navigation setup, where the inertial sensor would be co-located with the optical markers instead of with the camera.