Visual odometry uses image information to estimate the ego-motion of a camera and is a practical technique for navigation of autonomous vehicles. The paper presents a novel stereo-based ego-motion estimation method for this purpose. The proposed method employs a motion decoupling strategy that uses a set of distant feature points to estimate the cameras’ rotation, and a set of close feature points to estimate the translation. The set of feature points used for the motion estimation is carefully screened with bucketing plus aging techniques by considering their spatial distribution and temporal evolution. It has been found that the age of the tracked features have a significant impact on motion estimation. Older features generate less tracking error and should be selected for motion estimation. The proposed method is tested on the KITTI benchmark dataset and evaluated by comparing it with the existing visual odometry systems. The experimental results demonstrate that the proposed method can significantly improve the calculation efficiency and meanwhile guarantee a satisfactory accuracy.

Visual odometry estimates the ego-motion of an agent (e.g., vehicle and robot) using image information and is a key component for autonomous vehicles and robotics. This paper proposes a robust and precise method for estimating the 6-DoF ego-motion, using a stereo rig with optical flow analysis. An objective function fitted with a set of feature points is created by establishing the mathematical relationship between optical flow, depth and camera ego-motion parameters through the camera’s 3-dimensional motion and planar imaging model. Accordingly, the six motion parameters are computed by minimizing the objective function, using the iterative Levenberg–Marquard method. One of key points for visual odometry is that the feature points selected for the computation should contain inliers as much as possible. In this work, the feature points and their optical flows are initially detected by using the Kanade–Lucas–Tomasi (KLT) algorithm. A circle matching is followed to remove the outliers caused by the mismatching of the KLT algorithm. A space position constraint is imposed to filter out the moving points from the point set detected by the KLT algorithm. The Random Sample Consensus (RANSAC) algorithm is employed to further refine the feature point set, i.e., to eliminate the effects of outliers. The remaining points are tracked to estimate the ego-motion parameters in the subsequent frames. The approach presented here is tested on real traffic videos and the results prove the robustness and precision of the method.

Obstacle detection is the basis for the Advanced Driving Assistance System (ADAS) to take obstacle avoidance measures. However, it is a very essential and challenging task to detect unexpected obstacles on the road. To this end, an unexpected obstacle detection method based on computer vision is proposed. We first present two independent methods for the detection of unexpected obstacles: a semantic segmentation method that can highlight the contextual information of unexpected obstacles on the road and an open-set recognition algorithm that can distinguish known and unknown classes according to the uncertainty degree. Then, the detection results of the two methods are input into the Bayesian framework in the form of probabilities for the final decision. Since there is a big difference between semantic and uncertainty information, the fusion results reflect the respective advantages of the two methods. The proposed method is tested on the Lost and Found dataset and evaluated by comparing it with the various obstacle detection methods and fusion strategies. The results show that our method improves the detection rate while maintaining a relatively low false-positive rate. Especially when detecting unexpected long-distance obstacles, the fusion method outperforms the independent methods and keeps a high detection rate.

Obstacle detection is the primary task of the Advanced Driving Assistance System (ADAS). However, it is very difficult to achieve accurate obstacle detection in complex traffic scenes. To this end, this paper proposes an obstacle detection method based on the local spatial features of point clouds. Firstly, the local spatial point cloud of a superpixel is obtained through stereo matching and the SLIC image segmentation algorithm. Then, the probability of the obstacle in the corresponding area is estimated from the spatial feature information of the local plane normal vector and the superpixel point-cloud height, respectively. Finally, the detection results of the two methods are input into the Bayesian framework in the form of probabilities for the final decision. In order to describe the traffic scene efficiently and accurately, the detection results are further transformed into a multi-layer stixel representation. We carried out experiments on the KITTI dataset and compared several obstacle detection methods. The experimental results indicate that the proposed method has advantages in terms of its Pixel-wise True Positive Rate (PTPR) and Pixel-wise False Positive Rate (PFPR), particularly in complex traffic scenes, such as uneven roads.

Visual odometry estimates the ego-motion of a vehicle using only the input of a single or multiple cameras mounted on the vehicle. This paper focuses on the research of the stereo visual odometry system of intelligent vehicles, and discusses how to improve the robustness, accuracy and efficiency. A new robust estimation algorithm, Locally Optimized Progressive Sample Consensus algorithm, is proposed. Compared with the RANSAC algorithm, it can not only improve the accuracy of model estimation, but also can find more inliers to terminate the iteration process in advance, thereby speeding up the algorithm. A decoupling-based motion estimation algorithm is proposed. Monocular method is used to estimate the rotation parameters, which eliminates the influence of mismatching between left and right frames on rotation estimation. Moreover, when estimating the translational motion, the decoupling method makes the normalized re-projection error criterion better distinguish between inliers and outliers. The performance of the method is evaluated on KITTI benchmark dataset by comparing it with the existing visual odometry systems. The experimental results show that the proposed technique has a high accuracy and efficiency.