Abstract: Stereo matching algorithms that aim at high accuracy are not suitable for real-time applications on edge devices. To solve these problems, a fast stereo matching algorithm based on adaptive iterative residue optimization is proposed. This algorithm proposed a dual attention guided feature aggregation module to improve the algorithm's ability to represent features. An adaptive cross-cross matching module was proposed to improve the matching accuracy and robustness of the algorithm. An adaptive disparity range estimation module was proposed to reduce the error accumulation in iterations and reduce the amount of calculation. The evaluation results of three standard data sets show that the accuracy of the proposed algorithm is 1.55 percentage points higher than that of the algorithm with the same speed, and the speed of the proposed algorithm is 3 times higher than that of the algorithm with the same accuracy.