Advanced vehicle-based safety and warning systems use laser scanners to measure road geometry (position, curvature), and range to obstacles in order to warn a driver of an impending crash and/or to activate safety devices (air bags, brakes, steering). In order to objectively quantify performance of such a system, the reference system must be an order of magnitude more accurate than the sensors used by the warning system. This can be achieved by using high-resolution range images that can accurately perform object tracking and velocity estimation. Currently, this is very difficult to achieve when the measurements are taken from fast moving vehicles. Thus, our main objective is to improve motion estimation, which involves both rotational and translation movement of objects. In this respect, we present an innovative recursive motion estimation technique that can take advantage of the in depth resolution (range) to perform an accurate estimation of objects that have undergone 3-D translational and rotational movements. This approach iteratively aims at minimizing the error between the object in the current frame and its compensated object using estimated motion displacement from the previous range measurements. In addition, in order to use the range data on the non-rectangular grid in the Cartesian coordinate, two approaches have been considered: i) membrane fit, which interpolates the non-rectangular grid to rectangular grid, and ii) the non-rectangular grid range data by employing derivative filters and our proposed transformation between the Cartesian coordinates and the sensor-centered coordinates. For sequences of moving range images we demonstrate the effectiveness of the proposed scheme.
IEEE Transactions on Intelligent Transportation Systems
3-D motion estimation, range image, object tracking, ladar, laser scanners, vehicle safety, intelligent transport