Vehicular Edge Computing Based Driver Recommendation System Using Federated Learning
01 Dec 2020-pp 675-683
TL;DR: In this paper, the authors used Long Short-Term Memory Fully Convolutional Network to predict the corresponding stress level of the driver and established a relationship between stress and driving behavior and developed an intelligent recommendation system for cab companies to recommend the driver for a subsequent trip.
Abstract: Driver Stress and Behavior prediction is a significant feature of the Advanced Driver Assistance System. This system can improve driving safety by alerting the driver to the danger of unsafe or risky driving conditions. In this paper, we analyzed historical trip data to calculate the driving stress and its impact on different driving behavior. We used Long Short-Term Memory Fully Convolutional Network to predict the corresponding stress level of the driver. We further established a relationship between stress and driving behavior and developed an intelligent recommendation system for cab companies to recommend the driver for a subsequent trip. To meet the demand for Artificial Intelligence in the Intelligent Transportation System, we leverage Federated Learning in Vehicular Edge Computing in the proposed system architecture. It enables Road Side Units to do all computing of data on it. The model has been tested on the UAH-DriveSet dataset. We observed that the proposed model predicts the stress with an accuracy of 95% and assists in enhancing the driving quality and experience.
Citations
More filters
TL;DR: In this paper , a novel online computation offloading approach that could effectively reduce task latency and energy consumption in dynamic MEC with large-scale wireless user devices (WUDs) was proposed to accelerate the learning process.
Abstract: Mobile edge computing (MEC) has been envisioned as a promising paradigm that could effectively enhance the computational capacity of wireless user devices (WUDs) and quality of experience of mobile applications. One of the most crucial issues of MEC is computation offloading, which decides how to offload WUDs’ tasks to edge severs for further intensive computation. Conventional mathematical programming-based offloading approaches could face troubles in dynamic MEC environments due to the time-varying channel conditions (caused primarily by WUD mobility). To address the problem, reinforcement learning (RL) based offloading approaches have been proposed, which develop offloading policies by mapping MEC states to offloading actions. However, these approaches could fail to converge in large-scale MEC due to the exponentially-growing state and action spaces. In this article, we propose a novel online computation offloading approach that could effectively reduce task latency and energy consumption in dynamic MEC with large-scale WUDs. First, a RL-based computation offloading and energy transmission algorithm is proposed to accelerate the learning process. Then, a joint optimization method is adopted to develop the allocating algorithm, which obtains near-optimal solutions for energy and computation resources allocation. Simulation results show that the proposed approach can converge efficiently and achieve significant performance improvements over baseline approaches.
7 citations
TL;DR: This research summarizes the current limitations, highlights the areas that need improvements, and presents future paths for the development of robust federated recommenders that can handle the challenges of federated learning and, at the same time, generate high-quality recommendations.
5 citations
TL;DR: In this paper , the authors provide a detailed description of the critical technologies, challenges, and applications of federated learning in cloud-edge collaborative architecture, and provide guidance on future research directions.
Abstract: Abstract In recent years, with the rapid growth of edge data, the novel cloud-edge collaborative architecture has been proposed to compensate for the lack of data processing power of traditional cloud computing. On the other hand, on account of the increasing demand of the public for data privacy, federated learning has been proposed to compensate for the lack of security of traditional centralized machine learning. Deploying federated learning in cloud-edge collaborative architecture is widely considered to be a promising cyber infrastructure in the future. Although each cloud-edge collaboration and federated learning is hot research topic respectively at present, the discussion of deploying federated learning in cloud-edge collaborative architecture is still in its infancy and little research has been conducted. This article aims to fill the gap by providing a detailed description of the critical technologies, challenges, and applications of deploying federated learning in cloud-edge collaborative architecture, and providing guidance on future research directions.
4 citations
01 Jan 2022
TL;DR: inspired by the Federated Learning (FL) approach, this work tailor a collective perception architecture, introducing Situational Collective Perception (SCP) based on dynamically trained and situational DNNs, and enabling adaptive and efficient collective perception in future vehicular networks.
Abstract: : With the emerge of Vehicle-to-everything (V2X) communication, vehicles and other road users can perform Collective Perception (CP), whereby they exchange their individually detected environment to increase the collective awareness of the surrounding environment. To detect and classify the surrounding environmental objects, preprocessed sensor data (e.g., point-cloud data generated by a Lidar) in each vehicle is fed and classified by onboard Deep Neural Networks (DNNs). The main weakness of these DNNs is that they are commonly statically trained with context-agnostic data sets, limiting their adapt-ability to specific environments. This may eventually prevent the detection of objects, causing safety disasters. Inspired by the Federated Learning (FL) approach, in this work we tailor a collective perception architecture, introducing Situational Collective Perception (SCP) based on dynamically trained and situational DNNs, and enabling adaptive and efficient collective perception in future vehicular networks.
2 citations
TL;DR: In this paper , the authors introduced the concept of gradient encryption in federated learning (FL), which preserves the users' privacy without the additional computation requirements, and the computational power present in the edge devices helps to fine tune the local model and encrypt the input data to preserve privacy without any drop in performance.
Abstract: Autonomous vehicles (AVs) are getting popular because of their usage in a wide range of applications like delivery systems, self-driving taxis, and ambulances. AVs utilize the power of machine learning (ML) and deep learning (DL) algorithms to improve their self-driving learning experiences. The sudden surge in the number of AVs raises the need for distributed learning ecosystem to optimize their self-driving experiences at a rapid pace. Toward this goal, federated learning (FL) benefits, which can create a distributed learning environment for AVs. But, the traditional FL transfers the raw input data directly to a server, which leads to privacy concerns among the end-users. The concept of blockchain helps us to protect privacy, but it requires additional computational infrastructure. The extra infrastructure increases the operational cost for the company handling and maintaining the AVs. Motivated by this, in this paper, the authors introduced the concept of gradient encryption in FL, which preserves the users’ privacy without the additional computation requirements. The computational power present in the edge devices helps to fine-tune the local model and encrypt the input data to preserve privacy without any drop in performance. For performance evaluation, the authors have built a German traffic sign recognition system using a convolutional neural network (CNN) algorithm-based classification system and GeFL. The simulation process is carried out over a wide range of input parameters to analyze the performance at scale. Simulation results of GeFL outperform the conventional FL-based algorithms in terms of accuracy, i.e., 2% higher. Also, the amount of data transferred among the devices in the network is nearly three times less in GeFL compared to the traditional FL.
1 citations
References
More filters
28 Oct 2013
TL;DR: In this paper, a real world driving experiment with 10 participants measuring a variety of physiological data as well as a post-hoc video rating session was conducted to analyze the differences in the workload in terms of road type and especially important parts of the route such as exits and on-ramps.
Abstract: Driving a car is becoming increasingly complex. Many new features (e.g., for communication or entertainment) that can be used in addition to the primary task of driving a car increase the driver's workload. Assessing the driver's workload, however, is still a challenging task. A variety of means are explored which rather focus on experimental conditions than on real world scenarios (e.g., questionnaires). We focus on physiological data that may be assessed in an non-obtrusive way in the future and is therefore applicable in the real world.Hence, we conducted a real world driving experiment with 10 participants measuring a variety of physiological data as well as a post-hoc video rating session. We use this data to analyze the differences in the workload in terms of road type as well as especially important parts of the route such as exits and on-ramps. Furthermore, we investigate the correlation between the objective assessed and subjective measured data.
101 citations
01 Oct 2017
TL;DR: The proposed Stacked-LSTM model has achieved state-of-the-art results on the UAH-DriveSet with much higher true positive rate as well as lower false positive rate in comparison to the baseline approach.
Abstract: We are presenting a novel approach for the task of driving behavior classification based on stacked LSTM Recurrent Neural Networks. Given a nine different sensor data captured using a smart phone internal sensors during a naturalistic driving sessions, we formulated the driving behavior classification problem as time-series classification task. Whereas, given a window sequence of fused feature vectors of sensor data at any time step of a driving trip, we can accurately classify the driving behavior during that window sequence from a three distinctive driving behavior classes, namely normal, aggressive or drowsy driving. We evaluated our proposed Stacked-LSTM model on one of the recent naturalistic driving behavior analysis and classification dataset, UAH-DriveSet. Our proposed Stacked-LSTM model has achieved state-of-the-art results on the UAH-DriveSet with much higher true positive rate as well as lower false positive rate in comparison to the baseline approach. We have also compared the performance of our proposed Stacked-LSTM model against a number of the common classification algorithms used in the driving behavior classification and analysis studies and we achieved F1-measure score of 91% with an improvement of more than 10% over the closest compared approaches.
96 citations
TL;DR: A simple method for the detection of the stress using only three time-domain features of the heart rate signal, which can be efficiently implemented on mobile devices is developed.
Abstract: The automated detection of stress is a central problem for ambient assisted living solutions. The paper presents the concepts and results of two studies targeted at stress detection with a low cost heart rate sensor, a chest belt. In the device validation study ( ), we compared heart rate data and other features from the belt to those measured by a gold standard device to assess the reliability of the sensor. With simple synchronization and data cleaning algorithm, we were able to select highly (>97%) correlated, low average error (2.2%) data segments of considerable length from the chest data for further processing. The protocol for the clinical study ( ) included a relax phase followed by a phase with provoked mental stress, 10 minutes each. We developed a simple method for the detection of the stress using only three time-domain features of the heart rate signal. The method produced accuracy of 74.6%, sensitivity of 75.0%, and specificity of 74.2%, which is impressive compared to the performance of two state-of-the-art methods run on the same data. Since the proposed method uses only time-domain features, it can be efficiently implemented on mobile devices.
73 citations
TL;DR: This paper introduces two kinds of machine learning methods for evaluating the fuel efficiency of driving behavior using the naturalistic driving data and shows that the proposed method can effectively identify the relationship between the driving behavior and the fuel consumption on both macro and micro levels, allowing for end-to-end fuel consumption feature prediction.
Abstract: Driving behavior has a large impact on vehicle fuel consumption. Dedicated study on the relationship between the driving behavior and fuel consumption can contribute to decreasing the energy cost of transportation and the development of the behavior assessment technology for the ADAS system. Therefore, it is vital to evaluate this relationship in order to develop more ecological driving assistance systems and improve the vehicle fuel economy. However, modeling driving behavior under the dynamic driving conditions is complex, making a quantitative analysis of the relationship between the driving behavior and the fuel consumption difficult. In this paper, we introduce two kinds of machine learning methods for evaluating the fuel efficiency of driving behavior using the naturalistic driving data. In the first stage, we use an unsupervised spectral clustering algorithm to study the macroscopic relationship between driving behavior and fuel consumption, using the data collected during the natural driving process. In the second stage, the dynamic information from the driving environment and natural driving data is integrated to generate a model of the relationship between various driving behaviors and the corresponding fuel consumption features. The dynamic environment factors are coded into a processable, digital form using a deep learning-based object detection method so that the environmental data can be linked with the vehicle's operating signal data to provide the training data for the deep learning network. The training data are labeled according to its fuel consumption feature distribution, which is obtained from the road segment data and historical driving data. This deep learning-based model can then be used as a predictor of the fuel consumption associated with different driving behaviors. Our results show that the proposed method can effectively identify the relationship between the driving behavior and the fuel consumption on both macro and micro levels, allowing for end-to-end fuel consumption feature prediction, which can then be applied in the advanced driving assistance systems.
66 citations
TL;DR: Experiments show that the multi-channel SOBI presents a promising framework to accurately detect drowsiness by merging multi-physiological information in a less complex way.
Abstract: It is well known that blink, yawn, and heart rate changes give clue about a human's mental state, such as drowsiness and fatigue. In this paper, image sequences, as the raw data, are captured from smart phones which serve as non-contact optical sensors. Video streams containing subject's facial region are analyzed to identify the physiological sources that are mixed in each image. We then propose a method to extract blood volume pulse and eye blink and yawn signals as multiple independent sources simultaneously by multi-channel second-order blind identification (SOBI) without any other sophisticated processing, such as eye and mouth localizations. An overall decision is made by analyzing the separated source signals in parallel to determine the driver's driving state. The robustness of the proposed method is tested under various illumination contexts and a variety of head motion modes. Experiments on 15 subjects show that the multi-channel SOBI presents a promising framework to accurately detect drowsiness by merging multi-physiological information in a less complex way.
62 citations