Search Results

Parking-slot detection plays a critical role in the self-parking system for autonomous driving. To enhance the complexity of the environmental situations in parking-slot datasets and reduce the difficulty of manual annotation, we design several data synthesis methods to generate new parking-slots under different situations. Methods introduced in this paper include synthesizing parking-slots in AVM (around view monitor) images, generating parking-slots in fisheye images and adding 2D symbols inside parking-slots to form special ones. To test the influence of our synthetic data, we conduct a series of experiments on different tasks. In the parking-slot detection experiments, we design a novel two-stage parking-slot detection method. We use YOLOv7 as the object detector and different from previous methods, we detect the complete parking-slots and marking points at the same time. Then we match marking points and give them a certain order in the second stage.

The development of autonomous driving generally requires enormous annotated data as training input. The availability and quality of annotated data have been major restrictions in industry. Data synthesis techniques are then being developed to generate annotated data. This paper proposes a 2D data synthesis pipeline using original background images and target templates to synthesize labeled data for model training in autonomous driving. The main steps include: acquiring templates from template libraries or alternative approaches, augmenting the obtained templates with diverse techniques, determining the positioning of templates in images, fusing templates with background images to synthesize data, and finally employing the synthetic data for subsequent detection and segmentation tasks. Specially, this paper synthesizes traffic data such as traffic signs, traffic lights, and ground arrow markings in 2D scenes based on the pipeline.

With the rapid development of autonomous driving technologies, the proportion of autonomous vehicles (AVs) will increase and influence road capacity. In this study, the simulation of mixed traffic flow was studied using an improved cellular automata model. Safety inter-vehicle spacing, the length of vehicles and reaction time are introduced into the cellular automata model. We delete the acceleration, deceleration and randomization rule for ideal conditions. Numerical simulations are utilized to analyze road capacity with different proportions of AVs. Road capacity is about 2200 pcu/h/lane for pure manual vehicle (MV) traffic flow and about 3600 pcu/h/lane for pure AV traffic flow. The capacity increases by 19.2% when there are 50% AVs in the traffic flow. And the capacity increases by around 63.6% due to the pure AV traffic flow.

One promising method for reducing fuel consumption and emissions, particularly in heavy duty trucks, is platooning. As the distance between vehicles decreases, the following vehicles will experience less aerodynamic drag on the front of the vehicle. However, reducing the velocity of the air contacting the front of the vehicle could have adverse effects on the temperature of the engine. To compensate for this effect, the energy consumption of the engine cooling system might increase, ultimately limiting the overall improvements obtained with platooning. Understanding the coupling between drag reduction and engine cooling load requirement is key for successfully implementing platooning strategies. Additionally, in a Connected and Automated Vehicle (CAV) environment, where information of the future engine load becomes available, the operation of the cooling system can be optimized in order to achieve the maximum fuel consumption reduction.