论文部分内容阅读
针对行星车软着陆过程中,视觉导航系统参数变化后星面再标定难度大的问题,在传统立体校正算法基础上提出了一种适用于行星车视觉导航的弱标定立体校正算法。算法利用地面标定结果和对应匹配点进行优化计算,约束校正矩阵形式,保证了校正结果为以光心为投影点的单应变换,并进一步简化参数,降低计算复杂度。在理论上证明了算法的正确性,实验验证了算法的有效性。实验结果表明,算法校正结果与已知相机内、外参数的欧几里德校正方法性能相当,校正误差和校正畸变均较小,满足行星车视觉导航系统要求。
Aiming at the problem that the calibration of visual navigation system is difficult after recalibration of the visual navigation system during the soft landing of the planetary car, a weakly calibrated stereo calibration algorithm suitable for visual navigation of the satellite car is proposed based on the traditional stereo correction algorithm. The algorithm uses the ground calibration results and the corresponding matching points to optimize the calculation, and constrains the correction matrix to ensure that the correction result is a homography transformation with the optical center as the projection point, further simplifying the parameters and reducing the computational complexity. The correctness of the algorithm is proved in theory, and the validity of the algorithm is verified by experiments. The experimental results show that the proposed algorithm correlates well with the Euclidean correction method of known and internal parameters of the camera. The correction error and the correction distortion are both small, which meets the requirements of the visual navigation system for a planetary car.