Car Following Prediction Based on Support Vector Regression and Multi-adaptive Regression Spline by Considering Instantaneous Reaction Time

Pahlavani, Parham; Moghadam, Mohsen Poor Arab; Bigdeli, Behnaz

doi:10.1007/s40996-018-0141-0

Cited by 4 publications

(2 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Moreover, the motion of the vehicle introduces a Doppler shift which causes dynamic contraction of the visual zone [47], [48]. Furthermore, the solutions to the isolated sub-problems could be optimum, fine-tuned and well-solved, but might not integrate so as to result in a cohrerent solution.…”

Section: Empirical Decision-making System For Autonomous Vehiclesmentioning

confidence: 99%

Safe Driving Of Autonomous Vehicles Through Improved Deep Reinforcement Learning

Gupta¹

2021

Preprint

View full text Add to dashboard Cite

In this thesis, we propose an environment perception framework for autonomous driving using deep reinforcement learning (DRL) that exhibits learning in autonomous vehicles under complex interactions with the environment, without being explicitly trained on driving datasets. Unlike existing techniques, our proposed technique takes the learning loss into account under deterministic as well as stochastic policy gradient. We apply DRL to object detection and safe navigation while enhancing a self-driving vehicle’s ability to discern meaningful information from surrounding data. For efficient environmental perception and object detection, various Q-learning based methods have been proposed in the literature. Unlike other works, this thesis proposes a collaborative deterministic as well as stochastic policy gradient based on DRL. Our technique is a combination of variational autoencoder (VAE), deep deterministic policy gradient (DDPG), and soft actor-critic (SAC) that adequately trains a self-driving vehicle. In this work, we focus on uninterrupted and reasonably safe autonomous driving without colliding with an obstacle or steering off the track. We propose a collaborative framework that utilizes best features of VAE, DDPG, and SAC and models autonomous driving as partly stochastic and partly deterministic policy gradient problem in continuous action space, and continuous state space. To ensure that the vehicle traverses the road over a considerable period of time, we employ a reward-penalty based system where a higher negative penalty is associated with an unfavourable action and a comparatively lower positive reward is awarded for favourable actions. We also examine the variations in policy loss, value loss, reward function, and cumulative reward for ‘VAE+DDPG’ and ‘VAE+SAC’ over the learning process.

show abstract

Section: Empirical Decision-making System For Autonomous Vehiclesmentioning

confidence: 99%

Safe Driving Of Autonomous Vehicles Through Improved Deep Reinforcement Learning

Gupta¹

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…On the other hand, expansion of road networks as a solution to counter traffic congestion is not practical due to financial and environmental constraints (Knorr, Baselt et al 2012, Guo, Li et al 2019. Intelligent transportation systems are an effective approach for better managing traffic and related issues in smart cities (Mohan, Padmanabhan et al 2008;Pahlavani et al, 2019;Moghadam et al, 2017). Moreover, geospatial information systems provides a powerful tools for importing, analyzing, and displaying urban data from a variety of sensors and sources (Delavar, 2004;Pahlavani et al, 2006;Pahlavani et al, 2017;Bahari et al, 2014).…”

Section: Introductionmentioning

confidence: 99%

Optimizing Energy Consumption of Vehicle Sensor Networks Based on the K-Means Clustering Method and Ant Colony Algorithm

Mashari

Pahlavani

Ebrahimi

2019

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

With the world's growing population, the number of vehicles has increased, but the capacity of roads and transportation systems has not. The increasing traffic and the pollution that it causes has therefore become a problem all over the world. Wireless Sensor Networks that detect traffic and prevent its pernicious effects have attracted the attention of many. Fast information transfers, easy installation, less repair and maintenance, compression and lower costs make WSNs more common than other network solutions (Nellore and Hancke 2016). Since more traffic congestion wastes time and energy, it is crucial to develop and present approaches to more accurately detect traffic patterns. Clustering is one of the best data analysis methods used for detecting traffic patterns. The approach proposed by this study assumes that all vehicles are equipped with GPS, and that sensors establish connections with each other through radio communication equipment. In this case, clustering is used to gather important traffic information and create intelligent urban transportation systems (ITS) to reduce sensor energy consumption in vehicular sensor networks. The K-means method and the ANT Colony optimization algorithm were used to cluster sensors and investigate their impact on reducing sensor energy consumption. The results show that the K-means algorithm and the ANT Colony algorithm reduce vehicular sensor energy consumption by 41.7 and 76.8 percent, respectively. The investigations showed that the ANT Colony algorithm outperformed the Kmeans algorithm by 84.2%.

show abstract

Application of Multivariate Adaptive Regression Splines and Classification and Regression Trees to Estimate Wave-Induced Scour Depth Around Pile Groups

Samadi

Afshar

Jabbari

et al. 2020

Iran J Sci Technol Trans Civ Eng

View full text Add to dashboard Cite

Car Following Prediction Based on Support Vector Regression and Multi-adaptive Regression Spline by Considering Instantaneous Reaction Time

Cited by 4 publications

References 22 publications

Safe Driving Of Autonomous Vehicles Through Improved Deep Reinforcement Learning

Safe Driving Of Autonomous Vehicles Through Improved Deep Reinforcement Learning

Optimizing Energy Consumption of Vehicle Sensor Networks Based on the K-Means Clustering Method and Ant Colony Algorithm

Application of Multivariate Adaptive Regression Splines and Classification and Regression Trees to Estimate Wave-Induced Scour Depth Around Pile Groups

Contact Info

Product

Resources

About