A5001993723- Vehicular Ad Hoc Networks (VANETs)
- Robotics and Sensor-Based Localization
- Indoor and Outdoor Localization Technologies
- Target Tracking and Data Fusion in Sensor Networks
- Autonomous Vehicle Technology and Safety
- Privacy-Preserving Technologies in Data
- Robotics and Automated Systems
- Traffic control and management
- Advanced Neural Network Applications
- Robotic Path Planning Algorithms
- Mobile Crowdsensing and Crowdsourcing
- Human-Automation Interaction and Safety
- IoT and Edge/Fog Computing
- Ergonomics and Musculoskeletal Disorders
- Reinforcement Learning in Robotics
- Inertial Sensor and Navigation
- Fault Detection and Control Systems
- Gaussian Processes and Bayesian Inference
- Mobile Ad Hoc Networks
- Face recognition and analysis
- Brain Tumor Detection and Classification
- Ergonomics and Human Factors
- User Authentication and Security Systems
- Cognitive Computing and Networks
- Advanced Authentication Protocols Security
Industrial Systems Institute
2020-2025
University of Patras
2020-2024
Research Academic Computer Technology Institute
2021
Athena Research and Innovation Center In Information Communication & Knowledge Technologies
2021
Abstract The main goal of the H2020-CARAMEL project is to address cybersecurity gaps introduced by new technological domains adopted modern vehicles applying, among others, advanced Artificial Intelligence and Machine Learning techniques. As a result, CARAMEL enhances protection against threats related automated driving, smart charging Electric Vehicles, communication or between roadside infrastructure. This work focuses on latter presents architecture aiming at assessing integrity...
In this paper, we design distributed multi-modal localization approaches for Connected and Automated vehicles. We utilize information diffusion on graphs formed by moving vehicles, based Adapt-then-Combine strategies coupled with the Least-Mean-Squares Conjugate Gradient algorithms. treat vehicular network as an undirected graph, where vehicles communicate each other means of Vehicle-to-Vehicle communication protocols. perform cooperative fusion different measurement modalities, including...
Empowering each vehicle with four dimensional (4D) situational awareness, i.e., accurate knowledge of neighboring vehicles' 3D locations over time in a cooperative manner (instead focusing only on self-localization), is fundamental for improving autonomous driving performance diverse traffic conditions. For this task, identification, localization and tracking nearby road users critical enhancing safety, motion planning energy consumption automated vehicles. Advanced perception sensors as...
Cooperative Localization is expected to play a crucial role in various applications the field of Connected and Autonomous vehicles (CAVs). Future 5G wireless systems are enable cost-effective Vehicle-to-Everything (V2X) systems, allowing CAVs share with other entities network data they collect measure. Typical measurement models usually deployed for this problem, absolute position from Global Positioning System (GPS), relative distance azimuth angle neighbouring vehicles, extracted Light...
Cooperative Real-time Localization is expected to play a crucial role in various applications the field of Connected and Semi-Autonomous vehicles (CAVs), such as collision avoidance/warning, cooperative adaptive cruise control, etc. Future 5G wireless systems are enable cost-effective Vehicle-to-Everything (V2X) systems, allowing CAVs share measured data with other entities network. Typical measurement models usually deployed for this problem, absolute position from Global Positioning System...
Cooperative autonomous driving in 5G and smart cities environment is expected to further improve safety, security efficiency of transportation systems. To this end, involved vehicles imperative have accurate knowledge both their own neighboring vehicles' location, a task known as cooperative awareness. In paper, we formulated two novel distributed localization tracking schemes, based on Gradient Descent Extended Kalman Filter algorithms, cope with erroneous GPS location. Sensor-rich exploit...
With the advent of Connected and Autonomous Vehicles (CAVs) comes very real risk that these vehicles will be exposed to cyber-attacks by exploiting various vulnerabilities. This paper gives a technical overview H2020 CARAMEL project (currently in intermediate stage) which Artificial Intelligent (AI)-based cybersecurity for CAVs is main goal. Most possible scenarios are considered, an adversary can generate attacks on CAVs, such as camera sensors, GPS location, Vehicle Everything (V2X)...
The main goals of the CARAMEL project are to enhance protection modern vehicles against cybersecurity threats related automated driving, smart charging Electric Vehicles, and communication among or between roadside infrastructure. This work focuses on latter presents architecture for improving security privacy connected autonomous driving. proposed includes: (i) multi-radio access technology capabilities, with simultaneous 802.11p LTE-Uu support; (ii) a MEC platform, where algorithms...
Electric vehicles (EVs) gain great attention nowadays since the electrification of private and public transport has a potential to reduce greenhouse gas emissions mitigate oil dependency. However, influx large number electrical loads without any coordination could have adverse affects grid. More importantly, complexity in EVs, pose critical challenges ensuring overall system integrity. A typical attack found controllers connected EVs is false data injection (FDI), which can be utilized...
In this paper, we propose a novel cooperative-based system that facilitates each autonomous vehicle of the swarm to be fully aware its 5 degrees freedom (DOF) motion, i.e., 3D translation and 2D rotation, very important task for navigation, known also as simultaneous localization mapping (SLAM). The novelty is interconnected vehicles share common collective task: simultaneously estimating self neighboring vehicles' DOF by perceiving, transmitting, associating fusing heterogeneous data, e.g.,...
In this paper we investigate the impact of federated learning on deep learning-based feature extraction used for self-localization in autonomous vehicles. The accurate and reliable determination vehicle's position is crucial both safety efficiency purposes. Deep-learning based extractors have demonstrated benefits over model-based components started to replace them during past few years. addition, Federated Learning (FL) frameworks can provide advantages terms data privacy, resource...
Cooperative Localization has received extensive interest from several scientific communities including Robotics, Optimization, Signal Processing and Wireless Communications. It is expected to become a major aspect for number of crucial applications in the field Connected (Semi-) Autonomous vehicles (CAVs), such as collision avoidance/warning, cooperative adaptive cruise control, safely navigation, etc. 5G mobile networks will be key providing connectivity vehicle everything (V2X)...
Cooperative Intelligent Transportation Systems envision the integration of cooperative intelligence as a key operational part autonomous driving. In this way, fleet or swarm Connected and Automated Vehicles collectively coordinates its driving actions in order to maximize performance. To realize ambition, vehicles need be fully location-aware their surrounding environment, through distributed AI intelligence. Motivated by requirement, we develop paper awareness scheme which performs...
This paper proposes a novel localization framework based on collaborative training or federated learning paradigm, for highly accurate of autonomous vehicles. More specifically, we build the standard approach KalmanNet, recurrent neural network aiming to estimate underlying system uncertainty traditional Extended Kalman Filtering, and reformulate it by adapt-then-combine concept FedKalmanNet. The latter is trained in distributed manner group vehicles (or clients), with local datasets...
In this paper we propose a methodology combining Federated Learning (FL) with Cross-view Image Geo-localization (CVGL) techniques. We address the challenges of data privacy and heterogeneity in autonomous vehicle environments by proposing personalized scenario that allows selective sharing model parameters. Our method implements coarse-to-fine approach, where clients share only coarse feature extractors while keeping fine-grained features specific to local environments. evaluate our approach...
In collaborative tasks where humans work alongside machines, the robot's movements and behaviour can have a significant impact on operator's safety, health, comfort. To address this issue, we present multi-stereo camera system that continuously monitors posture while they with robot. This uses novel distributed fusion approach to assess in real-time help avoid uncomfortable or unsafe positions. The adjusts informs operator of any incorrect potentially harmful postures, reducing risk...