Rigid body motion from range image sequences

https://doi.org/10.1016/1049-9660(91)90001-6Get rights and content

Abstract

An algorithm is described for recovering the six degrees of freedom of motion of a vehicle from a sequence of range images of a static environment taken by a range camera rigidly attached to the vehicle. The technique utilizes a least-squares minimization of the difference between the measured rate of change of elevation at a point and the rate predicted by the so-called elevation rate constraint equation. It is assumed that most of the surface is smooth enough so that local tangent planes can be constructed, and that the motion between frames is smaller than the size of most features in the range image. This method does not depend on the determination of correspondences between isolated high-level features in the range images. The algorithm has been successfully applied to data obtained from the range imager on the Autonomous Land Vehicle (ALV). Other sensors on the ALV provide an initial approximation to the motion between frames. It was found that the outputs of the vehicle sensors themselves are not suitable for accurate motion recovery because of errors in dead reckoning resulting from such problems as wheel slippage. The sensor measurements are used only to approximately register range data. The algorithm described here then recovers the difference between the true motion and that estimated from the sensor outputs.

References (17)

  • B.K.P. Horn et al.

    Determining optical flow

    Artifi. Intelligence

    (1981)
  • M. Daily et al.

    Autonomous cross-country navigation with the ALV

  • R.E. Sampson

    3D range sensor-phase shift detection

    Computer

    (1987)
  • M. Hebert et al.

    3D vision for outdoor navigation by an autonomous vehicle

  • M. Asada

    Building a 3-D world model for a mobile robot from sensory data

  • D. Goldgof et al.

    Feature extraction and terrain matching

  • M. Daily et al.

    An operational perception system for cross-country navigation

  • B.K.P. Horn

    Automatic hill-shading and the reflectance map

There are more references available in the full text version of this article.

Cited by (65)

  • Direct hand pose estimation for immersive gestural interaction

    2015, Pattern Recognition Letters
    Citation Excerpt :

    In [17] Horn showed that it is only possible to have a direct solution for 3D motion estimation when some constraints are imposed, such as assuming the depth field is known, or assuming the motion is pure rotation or pure translation. Note that our new optical flow constraint equation (Eq. (6)) is different from the elevation rate constraint equation [18], although both the equations are based on the range images and are equivalent mathematically. When the elevation rate constraint equation is employed, one has to convert range images Z(x, y, t) captured from depth sensors to 3D Cartesian Elevation Map Z(X, Y, t).

  • Log-polar height maps for multiple range image registration

    2009, Computer Vision and Image Understanding
  • Ego-Motion Estimate Corruption Due to Violations of the Range Flow Constraint

    2018, IEEE International Conference on Intelligent Robots and Systems
  • A Brief History and Essentials of Bundle Adjustment

    2018, Wuhan Daxue Xuebao (Xinxi Kexue Ban)/Geomatics and Information Science of Wuhan University
View all citing articles on Scopus
View full text