Curved path tracking control is one of the most important functions of autonomous vehicles. First, small turning radius circular bends considering bend quadrant and travel direction restrictions are planned by polar coordinate equations. Second, an estimator of a vehicle state parameter and road adhesion coefficient based on an extended Kalman filter is designed. To improve the convenience and accuracy of the estimator, the combined slip theory, trigonometric function group fitting, and cubic spline interpolation are used to estimate the longitudinal and lateral forces of the tire model (215/55 R17). Third, to minimize the lateral displacement and yaw angle tracking errors of a four-wheel steering (4WS) vehicle, the front-wheel steering angle of the 4WS vehicle is corrected by a model predictive control (MPC) feed-back controller. Finally, CarSim® simulation results show that the 4WS autonomous vehicle based on the MPC feed-back controller can not only significantly improve the curved path tracking performance but also effectively reduce the probability of drifting or rushing out of the runway at high speeds and on low-adhesion roads.

No abstract

No abstract

In this paper, we propose the fast dense trajectories algorithm for human action recognition. Dense trajectories are robust to fast irregular motions and outperform other state-of-the-art descriptors such as KLT tracker or SIFT descriptors. However, the use of dense trajectories is time consuming. To improve the efficiency, we extract feature trajectories in the ROI rather than in the whole frames, and we use the temporal pyramids to achieve adaptable mechanism for different action speed. We evaluate the method on the dataset of Huawei/3DLife -3D human reconstruction and action recognition Grand Challenge in ACM Multimedia 2013. Experimental results show a significant improvement over the dense trajectories descriptor in real-time, and adaptable to different speed.

No abstract

No abstract

No abstract

No abstract