We integrate advanced computer vision and Vehicle-to-Infrastructure (V2I) communication systems for effective intersection navigation. In the first phase, the YOLOv8 deep learning model is employed to accurately detect traffic lights, with specialized training on the S2TLD Dataset for precision. Then we establish seamless V2I communication in MAVS Simulation, allowing vehicles to receive Signal Phase and Timing (SPaT) messages from traffic lights, enabling autonomous adjustment of speed and behavior. Simulating the scenarios in a high-fidelity automotive simulator demonstrates accurate traffic light detection and timely phase information, promising safer and more efficient intersection navigation for autonomous vehicles.
Adaptive Cruise Control (ACC) is one of the common advanced driving features that aims to assist the driver from fatigue. A lot of models for ACC are available, such as Proportional, Integral, and Derivative (PID), Model Predictive Control (MPC), and reinforcement learning (RL). Recently a deep RL technique called deep deterministic policy gradient (DDPG) has been widely used for velocity control. DDPG is a mixture of deep learning (DL) and RL method, that uses neural networks (NN) to estimate its action. In this work, we utilize an attention DDPG model. The attention mechanism in DDPG increases the overall effectiveness of the model by reducing focus on the less important features. The network structure consists of one hidden layer with 30 neurons, which has been deployed in our previous work. The reward function has been designed to concentrate on overall safety and comfort. In this paper, we introduce a new criterion of weather conditions, which has not been used before in the literature. We trained our model with a publicly available dataset. For testing, we created simulated sensor data created in the Mississippi State University Autonomous Vehicular Simulator (MAVS). We introduced various scenarios within the simulation environment with differing weather conditions to assess the performance of the ACC model. In many cases, the sensors are weather-sensitive, which can affect the performance of the ACC system. The primary objective of this study is to evaluate the performance of the Attention DDPG-based ACC model under varying weather conditions. The results show that the agent can maintain safety across a range of weather conditions.
For autonomous driving, pedestrian and road signs detection are key elements. There is much existing literature available addressing this issue successfully. However, the autonomous system requires a large and diverse set of training samples and labeling in real-world environments. Manual annotation of these samples is somewhat challenging and time-consuming. In this paper, our goal is to get better detection accuracy with minimal training data. For this, we have employed the active learning algorithm. Active learning is a useful method that selects only the effective portion of the dataset for training and reduces annotation costs. Though it uses only a small amount of the training data, it provides a high detection accuracy. In this work, we have chosen the deep active learning model for object detection via the probabilistic model of Choi et al. and modified the depth scale of different layers in the backbone. As real-world data may contain noise, motion, or other disruptions, we modified the original model to obtain improved detection results. In this experiment, we create a customized dataset that contains pedestrians, road signs, traffic lights, and zebra (or pedestrian) crossings to deploy the active learning algorithm. The experimental results show that the active learning model can produce good detection outcomes by accurately detecting and classifying pedestrians, road signs, traffic light, and zebra (or pedestrian) crossings.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.