ABSTRACT
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108 Multisensor Survey Vehicle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
Inertial Measuring Unit (IMU) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110 Forward-View and Side-View Cameras . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
Preprocessing of Raw Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 Inertial Navigation Fundamentals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 Estimation of Pose from Vision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
Extraction of Point Correspondents from a Sequence of Images . . . . . . . . 114 Determination of the Transformation between Vision-Inertial Sensors . . . . . . . 116
Optimization of the Vision-Inertial Transformation . . . . . . . . . . . . . . . . . . . . . 117 Sensor Fusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
IMU Error Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118 Design of the Kalman Filter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
Design of Vision Only Kalman Filter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 Design of Master Kalman Filter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 Experimental Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 Transformation between Inertial and Vision Sensors . . . . . . . . . . . . . . . . . . . . . . . . . . 122 Results of the Vision/IMU Integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125 Biographies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
Intermittent loss of the GPS signal is a common problem encountered in intelligent land navigation based on GPS integrated inertial systems. This issue emphasizes the need for an alternative technology that would ensure smooth and reliable inertial navigation during GPS outages. This paper presents the results of an effort where data from vision and inertial sensors are integrated. However, for such integration one has to first obtain the necessary navigation parameters from the available sensors. Due to the variety in the measurements, separate approaches have to be utilized in estimating the navigation parameters. Information from a sequence of images captured by a monocular camera attached to a survey vehicle at a maximum frequency of three frames per second was used in upgrading the inertial system installed in the same vehicle for its inherent error accumulation. Specifically, the rotations and translations estimated from point correspondences tracked through a sequence of images were used in the integration. Also a prefilter is utilized to smooth out the noise associated with the vision sensor (camera) measurements. Finally, the position locations based on the vision sensor are integrated with the inertial system in a decentralized format using a Kalman filter. The vision/inertial integrated position estimates are successfully compared with those from inertial/GPS system output. This successful comparison demonstrates that vision can be used successfully to supplement the inertial measurements during potential GPS outages.
