scholarly journals Deep Learning and YOLOv3 Systems for Automatic Traffic Data Measurement by Moving Car Observer Technique

2021 ◽  
Vol 6 (9) ◽  
pp. 134
Author(s):  
Marco Guerrieri ◽  
Giuseppe Parla
Keyword(s):  
Deep Learning ◽  
Traffic Flow ◽  
Traffic Data ◽  
Fundamental Interest ◽  
Traffic Conditions ◽  
Traffic Stream ◽  
First Results ◽  
Highway Facilities ◽  
Flow Variables ◽  
Good Agreement

Macroscopic traffic flow variables estimation is of fundamental interest in the planning, designing and controlling of highway facilities. This article presents a novel automatic traffic data acquirement method, called MOM-DL, based on the moving observer method (MOM), deep learning and YOLOv3 algorithm. The proposed method is able to automatically detect vehicles in a traffic stream and estimate the traffic variables flow q, space mean speed vs. and vehicle density k for highways in stationary and homogeneous traffic conditions. The first application of the MOM-DL technique concerns a segment of an Italian highway. In the experiments, a survey vehicle equipped with a camera has been used. Using deep learning and YOLOv3 the vehicles detection and the counting processes have been carried out for the analyzed highway segment. The traffic flow variables have been calculated by the Wardrop relationships. The first results demonstrate that the MOM and MOM-DL methods are in good agreement with each other despite some errors arising with MOM-DL during the vehicle detection step due to a variety of reasons. However, the values of macroscopic traffic variables estimated by means of the Drakes’ traffic flow model together with the proposed method (MOM-DL) are very close to those obtained by the traditional one (MOM), being the maximum percentage variation less than 3%.

scholarly journals Traffic Flow Variables Estimation: An Automated Procedure Based on Moving Observer Method. Potential Application for Autonomous Vehicles

2019 ◽  
Vol 20 (3) ◽  
pp. 205-214
Author(s):  
Marco Guerrieri ◽  
Giuseppe Parla ◽  
Raffaele Mauro
Keyword(s):  
Traffic Flow ◽  
Traffic Control ◽  
Autonomous Vehicles ◽  
Control Strategies ◽  
Digital Cameras ◽  
Mean Flow ◽  
Real Time Traffic ◽  
First Results ◽  
Technical Advances ◽  
Flow Variables

Abstract The estimation of traffic flow variables (flow, space mean speed and density) plays a fundamental role in highways planning and designing, as well as in traffic control strategies. Moving Observer Method (MOM) allows traffic surveys in a road, or in a road network. This paper proposes a novel automated procedure, called MOM-AP based on Moving Observer Method and Digital Image Processing (DIP) Technique able to automatically detect (without human observers) and calculate flow q, space mean speed vs and density k in case of stationary and homogeneous traffic conditions. In order to evaluate how reliable is the MOM-AP, an experiment has been carried out in a segment of one two-lane single carriageway road, in Italy. 30 datasets for the segment have been collected (in total 30 round trips). A comparative analysis between MOM-AP and traditional MOM has been carried out. First results show that the current MOM-AP algorithms underestimate the local mean flow variable values of around 10%. Nowadays MOM-AP may be implemented in smartphone apps. Instead, in the near future, it is realistic expecting the increase in the use of automated procedures for calculating the traffic flow variables (based on the “moving observer method”), due to the amount of sensors and digital cameras employed in the new autonomous vehicles (AVs). Considering such technical advances, the MOM-AP is a feasible model for real-time traffic analyses of road networks.

scholarly journals A Research of Traffic Prediction using Deep Learning Techniques

2019 ◽  
Vol 8 (9S2) ◽  
pp. 725-728
Keyword(s):  
Deep Learning ◽  
Traffic Flow ◽  
Transport Systems ◽  
Traffic Data ◽  
Intelligent Transport Systems ◽  
Intelligent Transport ◽  
Traffic Flow Prediction ◽  
Flow Prediction ◽  
Learning Techniques ◽  
Modern Technologies

Traffic data is very important in designing a smart city. Now –a day’smany intelligent transport systems use modern technologies to predict traffic flow, to minimize accidents on road, to predict speed of a vehicle and etc. The traffic flow prediction is an appealing study field. Many techniques of data mining are employed to forecast traffic. Deep learning techniques can be used with technological progress to prevent information from real time. Deep algorithms are discussed to forecast real-world traffic data. When traffic data becomes big data, some techniques to improve the accuracy of trafficprediction are also discussed.

Smart Driving System for Improving Traffic Flow

2017 ◽  
Vol 7 (7) ◽  
pp. 236
Author(s):  
Rajesh Kumar Gupta ◽  
L. N. Padhy ◽  
Sanjay Kumar Padhi
Keyword(s):  
Traffic Flow ◽  
Traffic Congestion ◽  
Urban Areas ◽  
Adaptive Cruise Control ◽  
Human Perception ◽  
Cruise Control ◽  
Driving System ◽  
V2v Communication ◽  
Traffic Conditions ◽  
Cooperative Adaptive Cruise Control

Traffic congestion on road networks is one of the most significant problems that is faced in almost all urban areas. Driving under traffic congestion compels frequent idling, acceleration, and braking, which increase energy consumption and wear and tear on vehicles. By efficiently maneuvering vehicles, traffic flow can be improved. An Adaptive Cruise Control (ACC) system in a car automatically detects its leading vehicle and adjusts the headway by using both the throttle and the brake. Conventional ACC systems are not suitable in congested traffic conditions due to their response delay.  For this purpose, development of smart technologies that contribute to improved traffic flow, throughput and safety is needed. In today’s traffic, to achieve the safe inter-vehicle distance, improve safety, avoid congestion and the limited human perception of traffic conditions and human reaction characteristics constrains should be analyzed. In addition, erroneous human driving conditions may generate shockwaves in addition which causes traffic flow instabilities. In this paper to achieve inter-vehicle distance and improved throughput, we consider Cooperative Adaptive Cruise Control (CACC) system. CACC is then implemented in Smart Driving System. For better Performance, wireless communication is used to exchange Information of individual vehicle. By introducing vehicle to vehicle (V2V) communication and vehicle to roadside infrastructure (V2R) communications, the vehicle gets information not only from its previous and following vehicle but also from the vehicles in front of the previous Vehicle and following vehicle. This enables a vehicle to follow its predecessor at a closer distance under tighter control.

scholarly journals First Results in Leak Localization in Water Distribution Networks using Graph-Based Clustering and Deep Learning

2020 ◽  
Vol 53 (2) ◽  
pp. 16691-16696
Author(s):  
Luis Romero ◽  
Joaquim Blesa ◽  
Vicenç Puig ◽  
Gabriela Cembrano ◽  
Carlos Trapiello
Keyword(s):  
Deep Learning ◽  
Water Distribution ◽  
Distribution Networks ◽  
Water Distribution Networks ◽  
First Results ◽  
Leak Localization ◽  
Graph Based Clustering

Surrounding Vehicles’ Contribution to Car-Following Models: Deep-Learning-Based Analysis

2021 ◽  
pp. 036119812110186
Author(s):  
Saeed Vasebi ◽  
Yeganeh M. Hayeri ◽  
Peter J. Jin
Keyword(s):  
Deep Learning ◽  
Traffic Flow ◽  
Short Term Memory ◽  
Data Availability ◽  
Car Following ◽  
Preceding Vehicle ◽  
Long Short Term Memory ◽  
The Right ◽  
Mean Square Errors ◽  
Deep Learning Model

Relatively recent increased computational power and extensive traffic data availability have provided a unique opportunity to re-investigate drivers’ car-following (CF) behavior. Classic CF models assume drivers’ behavior is only influenced by their preceding vehicle. Recent studies have indicated that considering surrounding vehicles’ information (e.g., multiple preceding vehicles) could affect CF models’ performance. An in-depth investigation of surrounding vehicles’ contribution to CF modeling performance has not been reported in the literature. This study uses a deep-learning model with long short-term memory (LSTM) to investigate to what extent considering surrounding vehicles could improve CF models’ performance. This investigation helps to select the right inputs for traffic flow modeling. Five CF models are compared in this study (i.e., classic, multi-anticipative, adjacent-lanes, following-vehicle, and all-surrounding-vehicles CF models). Performance of the CF models is compared in relation to accuracy, stability, and smoothness of traffic flow. The CF models are trained, validated, and tested by a large publicly available dataset. The average mean square errors (MSEs) for the classic, multi-anticipative, adjacent-lanes, following-vehicle, and all-surrounding-vehicles CF models are 1.58 × 10−3, 1.54 × 10−3, 1.56 × 10−3, 1.61 × 10−3, and 1.73 × 10−3, respectively. However, the results show insignificant performance differences between the classic CF model and multi-anticipative model or adjacent-lanes model in relation to accuracy, stability, or smoothness. The following-vehicle CF model shows similar performance to the multi-anticipative model. The all-surrounding-vehicles CF model has underperformed all the other models.

Design and Development of Logistics and Transport Route Control System

2011 ◽  
Vol 71-78 ◽  
pp. 4261-4264
Author(s):  
Ru Wang ◽  
San Yuan Tang ◽  
Wei Xin Sun
Keyword(s):  
Control System ◽  
Traffic Flow ◽  
Future Development ◽  
Design And Development ◽  
Urban Road ◽  
Traffic Conditions

According that town plan is mainly drawn with software AutoCAD, this article realizes to automatically select a shortest transport route on urban road and dynamically display traffic flow based on VC++ and ObjectARX and lays a foundation for future development taking traffic limit, traffic conditions and other complex conditions into account.

Sign in / Sign up

Export Citation Format

Share Document