Optimal motion estimation from visual and inertial measurements

Cameras and inertial sensors are good candidates to be deployed together for autonomous vehicle motion estimation, since each can be used to resolve the ambiguities in the estimated motion that results from using the other modality alone. We present an algorithm that computes optimal vehicle motion estimates by considering all of the measurements from a camera, rate gyro, and accelerometer simultaneously. Such optimal estimates are useful in their own right, and as a gold standard for the comparison of online algorithms. By comparing the motions estimated using visual and inertial measurements, visual measurements only, and inertial measurements only against ground truth, we show that using image and inertial data together can produce highly accurate estimates even when the results produced by each modality alone are very poor Our test datasets include both conventional and omnidirectional image sequences, and an image sequence with a high percentage of missing data.

[1]  M. Brooks,et al.  What value covariance information in estimating vision parameters? , 2001, Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001.

[2]  William H. Press,et al.  The Art of Scientific Computing Second Edition , 1998 .

[3]  Oleg A. Yakimenko,et al.  Application of nonlinear filtering to navigation system design using passive sensors , 2001 .

[4]  R Chellappa,et al.  Robust structure from motion estimation using inertial data. , 2001, Journal of the Optical Society of America. A, Optics, image science, and vision.

[5]  Martial Hebert,et al.  Experimental Comparison of Techniques for Localization and Mapping Using a Bearing-Only Sensor , 2000, ISER.

[6]  S. B. Kang,et al.  Recovering 3 D Shape and Motion from Image Streams using Non-Linear Least Squares , 1993 .

[7]  Richard Szeliski,et al.  Recovering 3D Shape and Motion from Image Streams Using Nonlinear Least Squares , 1994, J. Vis. Commun. Image Represent..

[8]  Takeo Kanade,et al.  An Iterative Image Registration Technique with an Application to Stereo Vision , 1981, IJCAI.

[9]  Richard Szeliski,et al.  Recovering 3D shape and motion from image streams using nonlinear least squares , 1993, Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.

[10]  F. A. Seiler,et al.  Numerical Recipes in C: The Art of Scientific Computing , 1989 .

[11]  Kenichi Kanatani,et al.  Do we really have to consider covariance matrices for image features? , 2001, Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001.

[12]  Philip F. Mclauchlan,et al.  The Variable State Dimension Filter applied to Surface-Based Structure from Motion , 1999 .

[13]  Camillo J. Taylor,et al.  Camera trajectory estimation using inertial sensor measurements and structure from motion results , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[14]  Stephen M. Rock,et al.  RELATIVE POSITION ESTIMATION FOR INTERVENTION-CAPABLE AUVS BY FUSING VISION AND INERTIAL MEASUREMENTS , .