Deep neural network dynamic traffic routing system for vehicles

Lamouik, Imad; Yahyaouy, Ali; Sabri, My Abdelouahed

doi:10.1109/isacv.2018.8354012

Cited by 13 publications

(3 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Chow et al [14] proposed an adaptive trafc control algorithm to help the drivers respond to the current trafc state and control settings and fnd the fastest route to the destination. Lamouik et al [15] proposed a dynamic route system based on deep convolutional neural networks, which provides fast routes between source and target points, efectively improving vehicle travel efciency and reducing red light waiting. Chen et al [16] proposed a daily dynamic learning and adjustment model with bounded rationality, where travelers can dynamically update their departure time and travel route using real-time trafc status information provided by navigation systems and past historical experience.…”

Section: Literature Reviewmentioning

confidence: 99%

“…Step 4: Detect change strength (CS). Update the number q of subroute population according to CS, the formula of CS is shown in equation (15). If CS is greater than the threshold value, merge it into other populations and q � q − 1; otherwise q � q + 1, so as to increase the exploration ability of the search space.…”

Section: Multiple Population Division Mechanismmentioning

confidence: 99%

See 1 more Smart Citation

Ecological and Real-Time Route Selection Method for Multiple Vehicles in Urban Road Network

Yan,

Tang,

Peng

et al. 2023

Journal of Advanced Transportation

View full text Add to dashboard Cite

Traffic congestion has been a hot topic of research in the field of intelligent transportation, which can be alleviated by efficient route navigation. Most of the existing route planning methods are non-negotiated algorithms, which do not take into account the route conflicts and collaborative relationships between multiple vehicles. Also, most negotiated algorithms have not been comprehensively considered dynamic route collaboration between vehicles, large-scale efficient computation, environmental pollution, etc. Therefore, an ecological multivehicle real-time route selection model (EMR2SM) for urban road networks is firstly proposed in this paper, which combines real-time traffic conditions of the road network with travel time, distance, and exhaust emissions as optimization indicators. In order to solve the large-scale computation problem of traditional negotiated algorithms, an adaptive multiswarm bee colony (AMSBC) algorithm is designed, which efficiently solves the multivehicle dynamic route selection problem. AMSBC searches the optimal route for each vehicle in parallel through multiple population division and self-adaption mechanism, to make multivehicle route selection reach Nash equilibrium. Compared with three non-negotiated optimization algorithms based on swarm technology, EMR2SM is verified by experiments that it improves the efficiency and accuracy of the optimal route selection for multiple vehicles and reduces vehicle emissions, which can effectively reduce traffic congestion and environmental pollution.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Section: Multiple Population Division Mechanismmentioning

confidence: 99%

Ecological and Real-Time Route Selection Method for Multiple Vehicles in Urban Road Network

Yan,

Tang,

Peng

et al. 2023

Journal of Advanced Transportation

View full text Add to dashboard Cite

show abstract

“…Moreover, RNN‐RBM owing to its deep and parallel architecture, outperforms Back Propagation Neural Network (BPNN) and SVM with at least 17% more accurate prediction wherein its execution time is only 3% of that for BPNN and 2.3% of that for SVM. Imad Lamouik et al 61 proposed a deep convolutional neural networks based path recommendation system that plans a smart path for the vehicle at intersection grids based on real‐time speed, spatiotemporal information of other vehicles, and state of traffic lights. The simulation results state that this model showed better routes with lower travel time and less red light stops for efficient vehicle routing.…”

Section: Taxonomymentioning

confidence: 99%

Machine learning‐based traffic scheduling techniques for intelligent transportation system: Opportunities and challenges

Nama

Nath

Bechra

et al. 2021

Int J Communication

View full text Add to dashboard Cite

Summary The rise in traffic congestion has become a significant concern in the urban city environment. The conventional traffic control systems with inefficient human resources management fail to control the traffic discipline, leading to increased traffic density and road offenses. However, the intelligent transportation system (ITS) can provide safety, efficiency, and sustainability for large‐scale vehicular traffic dilemmas. ITS unites machine learning with the available traffic control force and performs real‐time police scheduling to ensure the smooth flow of traffic. Many researchers have demonstrated notable work in intelligent traffic police scheduling and deployment using various optimization algorithms. However, the compilation of such praiseworthy work as a whole is still missing. Motivated by these facts, we provide a comprehensive review of the machine learning‐based state‐of‐the‐art technologies that can be used to form a three‐tier solution taxonomy. The first tier describes various tools and technologies that can be utilized to collect traffic data. The second tier highlights the machine learning algorithms and their accuracy, which forms a pattern in the collected data and then yields some crucial information about traffic flow, congestion levels, and so forth. In the third tier, the most vital taxonomy layer, various traffic police scheduling strategies are discussed. The proposed survey also presents the use of cases of traffic police scheduling that elaborates this review's applicability in various domains. Finally, some of the key challenges in the subject being reviewed are discussed, which initiates a further scope of improvement.

show abstract

Model predictive-based DNN control model for automated steering deployed on FPGA using an automatic IP generator tool

Reda,

Bouzid,

Zghaibe

et al. 2024

Des Autom Embed Syst

View full text Add to dashboard Cite

With the increase in the non-linearity and complexity of the driving system’s environment, developing and optimizing related applications is becoming more crucial and remains an open challenge for researchers and automotive companies alike. Model predictive control (MPC) is a well-known classic control strategy used to solve online optimization problems. MPC is computationally expensive and resource-consuming. Recently, machine learning has become an effective alternative to classical control systems. This paper provides a developed deep neural network (DNN)-based control strategy for automated steering deployed on FPGA. The DNN model was designed and trained based on the behavior of the traditional MPC controller. The performance of the DNN model is evaluated compared to the performance of the designed MPC which already proved its merit in automated driving task. A new automatic intellectual property generator based on the Xilinx system generator (XSG) has been developed, not only to perform the deployment but also to optimize it. The performance was evaluated based on the ability of the controllers to drive the lateral deviation and yaw angle of the vehicle to be as close as possible to zero. The DNN model was implemented on FPGA using two different data types, fixed-point and floating-point, in order to evaluate the efficiency in the terms of performance and resource consumption. The obtained results show that the suggested DNN model provided a satisfactory performance and successfully imitated the behavior of the traditional MPC with a very small root mean square error (RMSE = 0.011228 rad). Additionally, the results show that the deployments using fixed-point data greatly reduced resource consumption compared to the floating-point data type while maintaining satisfactory performance and meeting the safety conditions

show abstract

Deep neural network dynamic traffic routing system for vehicles

Cited by 13 publications

References 14 publications

Ecological and Real-Time Route Selection Method for Multiple Vehicles in Urban Road Network

Ecological and Real-Time Route Selection Method for Multiple Vehicles in Urban Road Network

Machine learning‐based traffic scheduling techniques for intelligent transportation system: Opportunities and challenges

Model predictive-based DNN control model for automated steering deployed on FPGA using an automatic IP generator tool

Contact Info

Product

Resources

About