DeepTrack: Lightweight Deep Learning for Vehicle Trajectory Prediction in Highways

Vinit, Katariya,; Baharani, Mohammadreza; Morris, Nichole L.; Shoghli, Omidreza; Tabkhi, Hamed

doi:10.1109/tits.2022.3172015

Cited by 22 publications

(4 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…If the recalculated arrival times at the road sections are too different from the original predicted times, the module sends the updated times to the RSUs that manage the road sections of the traveled route. Given the effectiveness of machine learning in predicting road travel times in the literature [ 58 , 59 ], the support vector regression (SVR) technique [ 60 ] is used in this paper to predict the arrival time at each road section of a traveled route based on the relevant information.…”

Section: Research Methodology and Steps Of The Studymentioning

confidence: 99%

An Adaptive Bandwidth Management Algorithm for Next-Generation Vehicular Networks

Huang,

Hu,

Cheng

2023

Sensors

View full text Add to dashboard Cite

The popularity of video services such as video call or video on-demand has made it impossible for people to live without them in their daily lives. It can be anticipated that the explosive growth of vehicular communication owing to the widespread use of in-vehicle video infotainment applications in the future will result in increasing fragmentation and congestion of the wireless transmission spectrum. Accordingly, effective bandwidth management algorithms are demanded to achieve efficient communication and stable scalability in next-generation vehicular networks. To the best of our current knowledge, a noticeable gap remains in the existing literature regarding the application of the latest advancements in network communication technologies. Specifically, this gap is evident in the lack of exploration regarding how cutting-edge technologies can be effectively employed to optimize bandwidth allocation, especially in the realm of video service applications within the forthcoming vehicular networks. In light of this void, this paper presents a seamless integration of cutting-edge 6G communication technologies, such as terahertz (THz) and visible light communication (VLC), with the existing 5G millimeter-wave and sub-6 GHz base stations. This integration facilitates the creation of a network environment characterized by high transmission rates and extensive coverage. Our primary aim is to ensure the uninterrupted playback of real-time video applications for vehicle users. These video applications encompass video conferencing, live video, and on-demand video services. The outcomes of our simulations convincingly indicate that the proposed strategy adeptly addresses the challenge of bandwidth competition among vehicle users. Moreover, it notably boosts the efficient utilization of bandwidth from less crowded base stations, optimizes the fulfillment of bandwidth prerequisites for various video applications, and elevates the overall video quality experienced by users. Consequently, our findings serve as a successful validation of the practicality and effectiveness of the proposed methodology.

show abstract

Section: Research Methodology and Steps Of The Studymentioning

confidence: 99%

An Adaptive Bandwidth Management Algorithm for Next-Generation Vehicular Networks

Huang,

Hu,

Cheng

2023

Sensors

View full text Add to dashboard Cite

show abstract

“…In view of the time-varying and nonlinear nature of vehicle speed, a CNN-based architecture with two-channel input was proposed for predicting short-term speed [ 27 ]. Katariya et al [ 28 ] designed the temporal convolutional networks (TCNs) for VSP, which can provide more robust time prediction with less computation compared with traditional CNNs. In [ 1 ], to overcome the limitations of the single prediction method, a short-term traffic speed prediction model was presented by combining an improved TCN and graph convolution network (GCN).…”

Section: Related Workmentioning

confidence: 99%

A Double-Layer Vehicle Speed Prediction Based on BPNN-LSTM for Off-Road Vehicles

Liu

Liang

Chen

et al. 2023

Sensors

View full text Add to dashboard Cite

The accurate prediction of vehicle speed is crucial for the energy management of vehicles. The existing vehicle speed prediction (VSP) methods mainly focus on road vehicles and rarely on off-road vehicles. In this paper, a double-layer VSP method based on backpropagation neural network (BPNN) and long short-term memory (LSTM) for off-road vehicles is proposed. First of all, considering the motion characteristics of off-road vehicles, the VSP problem is established and the relationship between the variables in the problem is carefully analyzed. Then, the double-layer VSP framework is presented, which consists of speed prediction and information update layers. The speed prediction layer established by using LSTM is to predict vehicle speed in the horizon, and the information update layer built by BPNN is to update the prediction information. Finally, with the help of mining truck and loader operation scenarios, the proposed VSP method is compared with the analytical method, BPNN prediction method, and recurrent neural network (RNN) prediction method in terms of speed prediction accuracy. The results show that, under the premise of ensuring the real-time prediction performance, the average prediction error of the proposed BPNN-LSTM prediction method under two operation scenarios reduces by 48.14%, 35.82% and 30.09% compared with the other three methods, respectively. The proposed speed prediction method provides a new solution for predicting the speed of off-road vehicles, effectively improving the speed prediction accuracy.

show abstract

“…In [28], a three-channel framework with a heterogeneous edge-enhanced graph attention network is proposed to deal with the inherent heterogeneity of the different vehicles in a given scene. DeepTrack [12] introduces temporal and depthwise convolutions to provide a more robust encoding of vehicle dynamics while reducing computation and parameters, resulting in a faster, lighterweight network with competitive accuracy. [31] proposed recently, utilizes a graph-based spatial-temporal convolutional network and a gated recurrent unit to predict future vehicle paths.…”

Section: Vehicle Bird's-eye View Path Predictionmentioning

confidence: 99%

“…Social-GAN (V) [11] Social-GAN (PV) [11] Next [22] Multiverse [21] SimAug [20] ST against best in class models [7,12,19,32,31,23]. Reporting RMSE is crucial in vehicle path prediction as autonomous vehicles evaluate the future trajectories of surrounding subjects for every second in the future upto few seconds.…”

Section: Social-lstm [1]mentioning

confidence: 99%

Pishgu: Universal Path Prediction Architecture through Graph Isomorphism and Attentive Convolution

Noghre¹,

Vinit²,

Pazho³

et al. 2022

Preprint

View full text Add to dashboard Cite

Path prediction is an essential task for several real-world real-time applications, from autonomous driving and video surveillance to environmental monitoring. Most existing approaches are computation-intensive and only target a narrow domain (e.g., a specific point of view for a particular subject). However, many real-time applications demand a universal path predictor that can work across different subjects (vehicles, pedestrians), perspectives (bird's-eye, highangle), and scenes (sidewalk, highway). This article proposes Pishgu, a universal graph isomorphism approach for attentive path prediction that accounts for environmental challenges. Pishgu captures the inter-dependencies within the subjects in each frame by taking advantage of Graph Isomorphism Networks. In addition, an attention module is adopted to represent the intrinsic relations of the subjects of interest with their surroundings. We evaluate the adaptability of our approach to multiple publicly available vehicle (bird's-eye view) and pedestrian (bird's-eye and high-angle view) path prediction datasets. Pishgu's universal solution outperforms existing domain-focused methods by producing state-of-the-art results for vehicle bird's-eye view by 42% and 61% and pedestrian high-angle views by 23% and 22% in terms of ADE and FDE, respectively. Moreover, we analyze the domain-specific details for various datasets to understand their effect on path prediction and model interpretation. Although our model is a single solution for path prediction problems and defines a new standard in multiple domains, it still has a comparable complexity to state-ofthe-art models, which makes it suitable for real-world application. We also report the latency and throughput for all three domains on multiple embedded processors.

show abstract

DeepTrack: Lightweight Deep Learning for Vehicle Trajectory Prediction in Highways

Cited by 22 publications

References 38 publications

An Adaptive Bandwidth Management Algorithm for Next-Generation Vehicular Networks

An Adaptive Bandwidth Management Algorithm for Next-Generation Vehicular Networks

A Double-Layer Vehicle Speed Prediction Based on BPNN-LSTM for Off-Road Vehicles

Pishgu: Universal Path Prediction Architecture through Graph Isomorphism and Attentive Convolution

Contact Info

Product

Resources

About