Explore the vast range of titles available.

Advancement in vision‐based techniques has enabled the autonomous vehicle system (AVS) to understand the driving scene in depth. The capability of autonomous vehicle system to understand the scene, and detecting the specific object depends on the strong feature representation of such objects. However, pothole objects are difficult to identify due to their non‐uniform structure in challenging, and dynamic road environments. Existing approaches have shown limited performance for the precise detection of potholes. The study on the detection of potholes, and intelligent driving behaviour of autonomous vehicle system is little explored in existing articles. Hence, here, an improved prototype model, which is not only truly capable of detecting the potholes but also shows its intelligent driving behaviour when any pothole is detected, is proposed. The prototype is developed using a convolutional neural network with a vision camera to explore, and validates the potential, and autonomy of its driving behaviour in the prepared road environment. The experimental analysis of the proposed model on various performance measures have obtained accuracy, sensitivity, and F‐measure of 99.02%, 99.03%, and 98.33%, respectively, which are comparable with the available state‐of‐art techniques.

The camera is one of the important sensors to realise the intelligent driving environment. It can realise lane detection and tracking, obstacle detection, traffic sign detection, identification and discrimination and visual simultaneous localisation and mapping. The visual sensor model, quantity and installation location are different on different intelligent driving hardware experimental platform as well as the visual sensor information processing module, thus a number of intelligent driving system software modules and interfaces are different. In this study, the software architecture of the autonomous vehicle based on the driving brain is used to adapt to different types of visual sensors. The target segment is extracted by the image segmentation algorithm, and then the segmentation of the region of interest is carried out. According to the input feature calculation results, the obstacle search is done in the second segmentation region, the output of the accessible road area. As driving information is complete, the authors will increase or reduce one or more visual sensors, change the visual sensor model or installation location, which will no longer directly affect the intelligent driving decision, they make the multi-vision sensors adapted to the requirements of different intelligent driving hardware test platforms.

Radar is an important sensor to realise intelligent driving environment perception, enabling the detection of static obstacles and dynamic obstacles, and the tracking of a dynamic obstacle. The models, quantities, and installing location of the platform radar sensors as well as the information processing modules differ from each other on different intelligent driving testing platforms, resulting in different quantities and interfaces on the intelligent driving system. Here, the authors build the software architecture of intelligent driving vehicle based on driving brain which is used to adapt to different types of radar sensors and use the variable granularity road ownership radar for radar information fusion. Under the condition of complete driving information, increasing or reducing the number of radar sensors and changing the radar sensor model or installing location will not affect the intelligent driving decision directly. Therefore, the authors meet the demands of multi-radar sensor adapting to different intelligent driving hardware testing platforms.

Driver behaviour is an important factor in road safety. Computer vision techniques have been widely used to monitor the driver behaviour. The violation of privacy and the possibility of spoofing are two continuing challenges in camera‐based systems. To address these challenges, we propose an efficient approach to monitor and detect driver behaviour based on movement characteristics of the vehicle rather than the visual features of the driver. The main goal of this paper is to classify the driver behaviour into five classes: safe, distracted, aggressive, drunk, and drowsy driving. A lightweight 1D Convolutional Neural Network with high efficiency and low computational complexity is suggested to classify the driver behaviour. Experimental results confirm that our method could successfully classify behaviours of a driver with accuracy of 99.999%.

Many traffic management and control applications require highly complete and accurate data of traffic flow. However, because of various reasons such as sensor failure or transmission error, it is common that some traffic flow data are lost. As a result, various methods were proposed by using a wide spectrum of techniques to estimate missing traffic data in the last two decades. Generally, these missing data imputation methods can be categorised into three kinds: prediction methods, interpolation methods and statistical learning methods. To assess their performance, these methods are compared from different aspects in this paper, including reconstruction errors, statistical behaviours and running speeds. Results show that statistical learning methods are more effective than the other two kinds of imputation methods when data of a single detector is utilised. Among various methods, the probabilistic principal component analysis (PPCA) yields best performance in all aspects. Numerical tests demonstrate that PPCA can be used to impute data online before making further analysis (e.g. make traffic prediction) and is robust to weather changes.

Driving distraction detection can effectively prevent the occurrence of traffic accidents. Thus, monitoring a driver's state is very important for road safety. At present, most driving distraction detection methods focus on singular aspects, such as gaze distraction or hand distraction, and rarely focus on cognitive distraction as it is difficult to detect. This study proposed a driving distraction detection method based on gaze activity. A multi‐channel convolutional neural network was used to classify the driver's gaze area and calculate the gaze activity. In cognitive distraction, the driver's gaze activity is significantly lower than that in the normal driving state. The activity thresholds via experiments have been obtained and used to determine whether drivers were in a cognitively distracted state. Through experiments, the accuracy of this method reached 92.36%. To identify driving distractions more comprehensively, a gaze distraction algorithm based on the two‐second rule has been added to the method. The experiment demonstrated that our method in combination with this algorithm improved all of the indicators (when compared to only the gaze activity algorithm), and the accuracy rate increased to 95%.

The zero‐velocity update (ZUPT) method is an effective way to reduce accumulated velocity errors of pedestrian navigation systems (PNSs). For a typical scheme, a stance phase detection module based on a fixed threshold is used to trigger the ZUPT algorithm. However, the detector is not robust enough for dynamic gait speeds. The false detection will degrade the navigation performance. In this letter, to improve the stance phase detector, the adaptive zero‐velocity detection problem is cast under dynamic gait speeds as a sequential threshold of a traditional detector inferring problem and a zero‐velocity detection framework proposed by combining a deep neural network with a traditional binary gait phase detector. Sufficient experimental results show that the proposed method outperforms other discussed learning‐based methods taking into account the trade‐off among model performance, structure, and size. Compared with the traditional method with a fixed threshold, the real‐world high‐dynamic positioning experiments show that this proposed method reduces the root mean squared error (RMSE) of absolute distance error by 48.7%, RMSE of start‐end error by 12.5%, and average RMSE of position error by 19.2%.

With the rapid development of Information and Communication Technologies, vehicular networks that communicate with each other will have an innovative application in traffic safety and congestion. This study describes a preliminary study on an automated on-ramp merging control algorithm for vehicles on freeways under condition of Internet-connected vehicles. On the basis of vehicular operation characteristics during the merging process analysis, a cooperative driving algorithm based on Internet of vehicles was designed to achieve ramp merging without collision. Then two on-ramp merging cases, including one vehicle and two vehicles merging into the platoon on main lane, were discussed in detail. Simulation works were carried out and the results proved that the on-ramp merging algorithm was effective, but the vehicle following the leading vehicle on ramp lane is disturbed seriously by the leading vehicle. At the same time, the simulation results also showed the scenario that merging a platoon into the two vehicles on main lane affects the traffic flow more seriously than letting each individual vehicle on ramp lane consecutively to merge in between the two vehicles in the main lane under the same initial condition.

Deep reinforcement learning has achieved some remarkable results in self‐driving. There is quite a lot of work to do in the area of autonomous driving with high real‐time requirements because of the inefficiency of reinforcement learning in exploring large continuous motion spaces. A deep imitation reinforcement learning (DIRL) framework is presented to learn control policies of self‐driving vehicles, which is based on a deep deterministic policy gradient algorithm (DDPG) by vision. The DIRL framework comprises two components, the perception module and the control module, using imitation learning (IL) and DDPG, respectively. The perception module employs the IL network as an encoder which processes an image into a low‐dimensional feature vector. This vector is then delivered to the control module which outputs control commands. Meanwhile, the actor network of the DDPG is initialized with the trained IL network to improve exploration efficiency. In addition, a reward function for reinforcement learning is defined to improve the stability of self‐driving vehicles, especially on curves. DIRL is verified by the open racing car simulator (TORCS), and the results show that the correct control strategy is learned successfully and has less training time.

Parking is becoming an expensive resource in almost any major city in the world, and its limited availability is a concurrent cause of urban traffic congestion, and air pollution. In old cities, the structure of the public parking space is rigidly organised and often in the form of on-street public parking spots. Unfortunately, these public parking spots cannot be reserved beforehand during the pre-trip phase, and that often lead to a detriment of the quality of urban mobility. Addressing the problem of managing public parking spots is therefore vital to obtain environmentally friendlier and healthier cities. Recent technological progresses in industrial automation, wireless network, sensor communication along with the widespread of high-range smart devices and new rules concerning financial transactions in mobile payment allow the definition of new intelligent frameworks that enable a convenient management of public parking in urban area, which could improve sustainable urban mobility. In such a scenario, the proposed intelligent parking assistant (IPA) architecture aims at overcoming current public parking management solutions. This study discusses the conceptual architecture of IPA and the first prototype-scale simulations of the system.