Multimodal Feature-Assisted Continuous Driver Behavior Analysis and Solving for Edge-Enabled Internet of Connected Vehicles Using Deep Learning

The emerging technology of internet of connected vehicles (IoCV) introduced many new solutions for accident prevention and traffic safety by monitoring the behavior of drivers. In addition, monitoring drivers’ behavior to reduce accidents has attracted considerable attention from industry and academic researchers in recent years. However, there are still many issues that have not been addressed due to the lack of feature extraction. To this end, in this paper, we propose the multimodal driver analysis internet of connected vehicles (MODAL-IoCV) approach for analyzing drivers’ behavior using a deep learning method. This approach includes three consecutive phases. In the first phase, the hidden Markov model (HMM) is proposed to predict vehicle motion and lane changes. In the second phase, SqueezeNet is proposed to perform feature extraction from these classes. Lastly, in the final phase, tri-agent-based soft actor critic (TA-SAC) is proposed for recommendation and route planning, in which each driver is precisely handled by an edge node for personalized assistance. Finally, detailed experimental results prove that our proposed MODAL-IoCV method can achieve high performance in terms of latency, accuracy, false alarm rate, and motion prediction error compared to existing works.

Driver Response Time in Cut-Off Scenarios from the Second Strategic Highway Research Program Naturalistic Database

A cut-in or cut-off scenario involves a vehicle intruding into the path of another vehicle traveling in the same direction. These lane changes can lead to potentially dangerous situations, either a sideswipe or a rear-end crash. In this study, 552 cut-in events were analyzed, including four crash and 548 near-crash events from the Second Strategic Highway Research Program (SHRP-2) data set. Video and onboard-data-recorder data from the responding vehicle were used to analyze various factors associated with drivers’ responses. Driver response times were measured from three different event onsets, and the effects of different factors on the respective response times were measured. These factors included the behavior of the subject driver, the behavior of the intruding vehicle/principal other vehicle (POV), and different environmental and infrastructural factors. The results showed that drivers responded more slowly when the POV took longer to move laterally to the subject driver’s lane edge and faster when this time was short. Similarly, drivers responded faster to merging vehicles that started from a stop. Yet, response times were no different when the POV utilized a directional signal. These results point to a kinematic threshold involving lateral distance and lateral speed that best describes how drivers were triggered to respond. Drivers also responded faster near intersections, and at night. The results can be utilized to design crash mitigation systems in autonomous vehicles, as well as non-automated vehicles, to supplement human responses where their abilities may be lacking.

Characteristics of Heavy Vehicle Discretionary Lane Changing Based on Trajectory Data

A comprehensive analysis of the motivations, gap acceptance, duration, and speed adjustment of heavy vehicle lane changes (LC) is conducted in this paper. An rich data set containing 433 discretionary LC trajectories of heavy vehicles is used in this study and the data set is divided into two data sets based on the LC direction (LC to the left lane [LCLL] and LC to the right lane [LCRL]) for comparison. It is seen that LCLL and LCRL have significantly different motivations, which also results in different gap acceptance behavior. However, the LC direction does not significantly influence the LC duration. The navigation speed significantly influences the LC duration of heavy vehicles and the LC duration will decrease with the increase of speed, indicating the substantial impact of traffic conditions on LC duration. An obvious speed synchronization phenomenon is found in the process of LCLL, which is not the case in LCRL. The results of this study highlight the distinct characteristics of the LC of heavy vehicles and produce a better understanding of the lane-changing behaviors of heavy vehicles. The fitted distributions of LC duration and further investigation into gap acceptance behaviors may be used for microscopic traffic simulation and auto driving.

Vehicle Lane-Changes Trajectory Prediction Model Considering External Parameters

The ability to predict the motion of vehicles is essential for autonomous vehicles. Aiming at the problem that existing models cannot make full use of the external parameters including the outline of vehicles and the lane, we proposed a model to use the external parameters thoroughly when predicting the trajectory in the straight-line and non-free flow state. Meanwhile, dynamic sensitive area is proposed to filter out inconsequential surrounding vehicles. The historical trajectory of the vehicles and their external parameters are used as inputs. A shared Long Short-Term Memory (LSTM) cell is proposed to encode the explicit states obtained by mapping historical trajectory and external parameters. The hidden states of vehicles obtained from the last step are used to extract latent driving intent. Then, a convolution layer is designed to fuse hidden states to feed into the next prediction circle and a decoder is used to decode the hidden states of the vehicles to predict trajectory. The experiment result shows that the dynamic sensitive area can shorten the training time to 75.86% of the state-of-the-art work. Compared with other models, the accuracy of our model is improved by 23.7%. Meanwhile, the model's ability of anti-interference of external parameters is also improved.