In order to realize automatic, universal and accurate identification and guidance of unstructured roads for unmanned equipment in the field environment, this paper proposed a road guidance algorithm for field scenes based on graph reasoning model and intelligent optimization. Firstly, the method segmented the image into homogeneous superpixel blocks, and fused multi-features of the superpixel blocks to construct a training set. Improved the traditional Laplace support vector machine algorithm, combined the location information of superpixel blocks, dynamically selected superpixel seed blocks in road areas, and trained multi-class classifier regressors of superpixel blocks and consistency regressors of adjacent superpixels. Combined the regression values of two kinds of regressors, constructed the energy function of Markov random field and then used the standard graph cutting algorithm to iteratively obtain the minimized energy function to realize the initial road reasoning segmentation. Combined the initial segmentation results of roads, the objective function was constructed according to the constraints set by people′s intuitive perception of roads, and used the differential immune clonal evolution algorithm to intelligently optimize and extract the guide lines of roads. The data collected in Zhushan, Nanjing and DARPA grand challenge database were tested, and it compared the results qualitatively and quantitatively with those of classical algorithms. The results show that the extraction accuracy of the guide line of unstructured roads by this algorithm in the field environment is over 91.79%, compared with classical algorithms, the detection accuracy is increased by 48.1% and 35.5% respectively, and the processing efficiency of the algorithm is increased by 98.6 % and 97.8 % respectively, which balances the real-time performance and accuracy of detection and has a strong application prospect.