Simultaneous Localization and Mapping (SLAM) is one of the key issues for mobile robots to achieve true autonomy. The implementations of SLAM could rely on a variety of sensors. Among many types of them, the laser-based SLAM approach is widely used owing to its high accuracy, even in poor lighting conditions. However, when in structure-less environments, laser modules will fail due to a lack of sufficient geometric features. Besides, motion estimation by moving lidar has the problem of distortion since range measurements are received continuously. To solve these problems, we propose a tightly-coupled SLAM integrating LiDAR and an integrated navigation system (INS) for unmanned vehicle navigation in campus environments. On the basis of feature extraction, a constraint equation for inter-frame point cloud features is constructed, and the pose solution results of the INS are added as a priori data for inter-frame point cloud registration. The Levenberg-Marquardt nonlinear least square method is used to solve the constraint equation to obtain inter-frame pose relationships. Map matching and loop closure detection methods are used to optimize the odometer, and the optimal pose information is obtained. The proposed SLAM algorithm is evaluated by comparing with the classic open-source laser SLAM algorithms on the campus dataset. Experimental results demonstrate that our proposed algorithm has certain advantages in estimating the trajectory error of the unmanned vehicle and has higher mapping performance.