Search

Shared control structure is beneficial to steering controller design of intelligence vehicles, and human-machine goal consistency is a key prerequisite for shared control. However, the goal consistency is usually given and cannot be changed, and the steering controller in low goal consistency, which directly affect the vehicle performance in case of emergency, has not been sufficiently investigated. This paper proposes a shared steering controller for path-following task based on Nash game strategy and steer-by-wire system considering different human-machine goal consistency. The driver-automation interactive path-following task is modeled by non-cooperative MPC, and authority weight of lateral displacement is used to balance the control objectives of the driver and automation. Human-machine goal consistency is determined by the driver and the automation controller steering angle. Aimed at different goal consistencies, a continuous authority weight adjustment algorithm is designed to ensure correct path following. This is especially true in low consistency in this study, when four driving modes are given to meet the different demand for control power. Simulations and hard-in-loop tests are conducted to verify the proposed control algorithm and the results show that it can perform the path-following task irrespective of human-machine goal consistency. [ABSTRACT FROM AUTHOR]

Abstract: Recognizing traffic signs is a challenging problem; and it has captured the attention of the computer vision community for several decades. Essentially, traffic sign recognition is a multi-class classification problem that has become a real challenge for computer vision and machine learning techniques. Although many machine learning approaches are used for traffic sign recognition, they are primarily used for classification, not feature design. Identifying rich features using modern machine learning methods has recently attracted attention and has achieved success in many benchmarks. However these approaches have not been fully implemented in the traffic sign recognition problem. In this paper, we propose a new approach to tackle the traffic sign recognition problem. First, we introduce a new feature learning approach using group sparse coding. The primary goal is to exploit the intrinsic structure of the pre-learned visual codebook. This new coding strategy preserves locality and encourages similar descriptors to share similar sparse representation patterns. Second, we use a non-uniform quantization approach based on log-polar mapping. Using the log-polar mapping of the traffic sign image, rotated and scaled patterns are converted into shifted patterns in the new space. We extract the local descriptors from these patterns to learn the features. Finally, by evaluating the proposed approach using the German Traffic Sign Recognition Benchmark dataset, we show that the proposed coding strategy outperforms existing coding methods and the obtained results are comparable to the state-of-the-art. [Copyright &y& Elsevier]