Compact and low-cost devices are needed for autonomous driving to image and measure distances to objects 360-degree around. We have been developing an omnidirectional stereo camera exploiting two hyperbolic mirrors and a single set of a lens and sensor, which makes this camera compact and cost efficient. We establish a new calibration method for this camera considering higher-order radial distortion, detailed tangential distortion, an image sensor tilt, and a lens-mirror offset. Our method

more »

... es the calibration error by 6.0 and 4.3 times for the upper-and lower-view images, respectively. The random error of the distance measurement is 4.9% and the systematic error is 5.7% up to objects 14 meters apart, which is improved almost nine times compared to the conventional method. The remaining distance errors is due to a degraded optical resolution of the prototype, which we plan to make further improvements as future work. KEY WORDS: electronics and control, semiconductor camera, image processing / Stereo camera [E1] Ryota Our method reduces the root mean square of the calibration accuracy by 6.0 times and 4.3 times for upper-and lower-view images, respectively. The systematic error of distance is 5.7% up to objects 14 meters apart, which is improved almost nine times compared to that calculated by the previous method and virtually identical to the target value of 5%. The standard deviation is at most 4.9%. As future work, an improvement in the optical resolution can lead to further improvement in accuracy. We have demonstrated that our camera is capable of locating objects, such as pedestrians, other cars, and obstacles within 14 meters. Besides distance measurements described in this paper, object recognition is also needed for autonomous driving. Further improvements and optimizations to our tentative recognition system will make this camera practical as a cost-effective sensor for autonomous driving. This paper is written based on a proceeding presented at JSAE