Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
4.5
2.7
Journals
Applied Sciences
Special Issues
Intelligent Vehicles: Overcoming Challenges
share announcement

Intelligent Vehicles: Overcoming Challenges

A special issue of Applied Sciences (ISSN 2076-3417). This special issue belongs to the section "Robotics and Automation".

Deadline for manuscript submissions: closed (30 September 2021) | Viewed by 12305

Special Issue Editors

Dr. Angelos Amanatiadis

E-Mail Website
Guest Editor
Department of Production and Management Engineering, Democritus University of Thrace, 67132 Xanthi, Greece
Interests: robotic systems; computer vision; machine learning; real-time embedded systems
Special Issues, Collections and Topics in MDPI journals
Dr. Javier Ibanez-Guzman

E-Mail Website
Guest Editor
Renault, France
Interests: autonomous vehicles; robotics; localization; connected intelligent vehicles, safety verification

Special Issue Information

Dear Colleagues,

While in the past few years, significant progress has been achieved in the intelligent vehicles domain, there are currently still no driverless vehicles in general use, and the path to full autonomy has been proven to be very challenging. Driving is largely a social interaction task while operating within the randomness encountered in public road networks. It requires situation understanding and decision inference under different conditions. Progress in machine learning is enabling major advances in different system functions, though introducing additional complexity. The challenges include how to address edge cases, how to ensure and demonstrate system safety, how to embed system resilience to ensure vehicle operability, etc. More advanced perception, decision-making, planning, and control algorithms are needed to enable autonomous navigation within the complexity found in public roads. The processing of significantly increased sensor data requires higher performance computing and hardware/software co-design to make the necessary real-time driving decisions. Novel training, testing, and validation methods are also emerging to provide reliability guarantees in the presence of autonomous vehicle failures.

This Special Issue aims to provide academia and industry results from the latest research developments and solutions that can contribute to autonomous vehicles’ social acceptability. The emphasis is on emergent techniques and practices, the products of progress in understanding the issues related to interaction, trustworthiness, and validation gained from the field experimentation of different approaches.

Asst. Prof. Angelos Amanatiadis
Dr. Javier Ibanez-Guzman
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Applied Sciences is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • Real-world and synthetic datasets for machine learning and benchmarking
  • Perception, mapping, and localization of autonomous navigation
  • Situation awareness, decision-making planning, and control in complex environments
  • Cooperative vehicles, assisted navigation using V2X
  • System safety, fault-tolerance, self-awareness, trustworthiness of intelligent vehicles
  • Methods and tools for the validation, verification, safety demonstration, and vehicle performance assessment

Published Papers (4 papers)

Download All Papers
Order results
Result details
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:

Research

17 pages, 4769 KiB  
Article
Vehicle State Estimation Using Interacting Multiple Model Based on Square Root Cubature Kalman Filter
by Wan Wenkang, Feng Jingan, Song Bao and Li Xinxin
Appl. Sci. 2021, 11(22), 10772; https://0-doi-org.brum.beds.ac.uk/10.3390/app112210772 - 15 Nov 2021
Cited by 14 | Viewed by 1888
Abstract
The distributed drive arrangement form has better potential for cooperative control of dynamics, but this drive arrangement form increases the parameter acquisition workload of the control system and increases the difficulty of vehicle control accordingly. In order to observe the vehicle motion state [...] Read more.
The distributed drive arrangement form has better potential for cooperative control of dynamics, but this drive arrangement form increases the parameter acquisition workload of the control system and increases the difficulty of vehicle control accordingly. In order to observe the vehicle motion state accurately and in real-time, while reducing the effect of uncertainty in noise statistical information, the vehicle state observer is designed based on interacting multiple model theory with square root cubature Kalman filter (IMM-SCKF). The IMM-SCKF algorithm sub-model considers different state noise and measurement noise, and the introduction of the square root filter reduces the complexity of the algorithm while ensuring accuracy and real-time performance. To estimate the vehicle longitudinal, lateral, and yaw motion states, the algorithm uses a three degree of freedom (3-DOF) vehicle dynamics model and a nonlinear brush tire model, which is then validated in a Carsim-Simulink co-simulation platform for multiple operating conditions. The results show that the IMM-SCKF algorithm’s fusion output results can effectively follow the sub-model with smaller output errors, and that the IMM-SCKF algorithm’s results are superior to the traditional SCKF algorithm’s results. Full article
(This article belongs to the Special Issue Intelligent Vehicles: Overcoming Challenges)
►▼ Show Figures

Figure 1

13 pages, 1366 KiB  
Article
Control Design and Assessment for a Reversing Tractor–Trailer System Using a Cascade Controller
by Abdullah Aldughaiyem, Yasser Bin Salamah and Irfan Ahmad
Appl. Sci. 2021, 11(22), 10634; https://0-doi-org.brum.beds.ac.uk/10.3390/app112210634 - 11 Nov 2021
Cited by 3 | Viewed by 1853
Abstract
In recent years, control design for unmanned systems, especially a tractor–trailer system, has gained popularity among researchers. The emergence of such interest is caused by the potential reduction in cost and shortage of number of workers and labors. Two industries will benefit from [...] Read more.
In recent years, control design for unmanned systems, especially a tractor–trailer system, has gained popularity among researchers. The emergence of such interest is caused by the potential reduction in cost and shortage of number of workers and labors. Two industries will benefit from the advancements of these types of systems: agriculture and cargo. By using the unmanned tractor–trailer system, harvesting and cultivating plants will become a safe and easy task. It will also cause a reduction in cost which in turn reduces the price on the end consumers. On the other hand, by using the unmanned tractor–trailer system in the cargo industry, shipping cost and time for the item delivery will be reduced. The work presented in this paper focuses on the development of a path tracking and a cascaded controller to control a tractor–trailer in reverse motion. The path tracking controller utilizes the Frenet–Serret frame to control the kinematics of the tractor–trailer system on a desired path, while the cascade controller’s main objective is to stabilize the system and to perform commands issued by the path tracker. The controlled parameters in this proposed design are the lateral distance to a path, trailer’s heading angel, articulated angel, and articulated angle’s rate. The main goal of such controller is to follow a path while the tractor–trailer system is moving in reverse and controlling the stability of the articulated vehicle to prevent the occurrence of a jackknife incident (uncontrolled state). The proposed controller has been tested in a different scenario where a successful implementation has been shown. Full article
(This article belongs to the Special Issue Intelligent Vehicles: Overcoming Challenges)
►▼ Show Figures

Figure 1

28 pages, 5581 KiB  
Article
Real-Time Semantic Image Segmentation with Deep Learning for Autonomous Driving: A Survey
by Ilias Papadeas, Lazaros Tsochatzidis, Angelos Amanatiadis and Ioannis Pratikakis
Appl. Sci. 2021, 11(19), 8802; https://0-doi-org.brum.beds.ac.uk/10.3390/app11198802 - 22 Sep 2021
Cited by 19 | Viewed by 5916
Abstract
Semantic image segmentation for autonomous driving is a challenging task due to its requirement for both effectiveness and efficiency. Recent developments in deep learning have demonstrated important performance boosting in terms of accuracy. In this paper, we present a comprehensive overview of the [...] Read more.
Semantic image segmentation for autonomous driving is a challenging task due to its requirement for both effectiveness and efficiency. Recent developments in deep learning have demonstrated important performance boosting in terms of accuracy. In this paper, we present a comprehensive overview of the state-of-the-art semantic image segmentation methods using deep-learning techniques aiming to operate in real time so that can efficiently support an autonomous driving scenario. To this end, the presented overview puts a particular emphasis on the presentation of all those approaches which permit inference time reduction, while an analysis of the existing methods is addressed by taking into account their end-to-end functionality, as well as a comparative study that relies upon a consistent evaluation framework. Finally, a fruitful discussion is presented that provides key insights for the current trend and future research directions in real-time semantic image segmentation with deep learning for autonomous driving. Full article
(This article belongs to the Special Issue Intelligent Vehicles: Overcoming Challenges)
►▼ Show Figures

Figure 1

20 pages, 2127 KiB  
Article
Hybrid Representation of Sensor Data for the Classification of Driving Behaviour
by Michalis Savelonas, Ioannis Vernikos, Dimitris Mantzekis, Evaggelos Spyrou, Athanasia Tsakiri and Stavros Karkanis
Appl. Sci. 2021, 11(18), 8574; https://0-doi-org.brum.beds.ac.uk/10.3390/app11188574 - 15 Sep 2021
Cited by 9 | Viewed by 1782
Abstract
Monitoring driving behaviour is important in controlling driving risk, fuel consumption, and CO2 emissions. Recent advances in machine learning, which include several variants of convolutional neural networks (CNNs), and recurrent neural networks (RNNs), such as long short-term memory (LSTM) and gated recurrent [...] Read more.
Monitoring driving behaviour is important in controlling driving risk, fuel consumption, and CO2 emissions. Recent advances in machine learning, which include several variants of convolutional neural networks (CNNs), and recurrent neural networks (RNNs), such as long short-term memory (LSTM) and gated recurrent unit (GRU) networks, could be valuable for the development of objective and efficient computational tools in this direction. The main idea in this work is to complement data-driven classification of driving behaviour with rules derived from domain knowledge. In this light, we present a hybrid representation approach, which employs NN-based time-series encoding and rule-guided event detection. Histograms derived from the output of these two components are concatenated, normalized, and used to train a standard support vector machine (SVM). For the NN-based component, CNN-based, LSTM-based, and GRU-based variants are investigated. The CNN-based variant uses image-like representations of sensor measurements, whereas the RNN-based variants (LSTM and GRU) directly process sensor measurements in the form of time-series. Experimental evaluation on three datasets leads to the conclusion that the proposed approach outperforms a state-of-the-art camera-based approaches in distinguishing between normal and aggressive driving behaviour without using data derived from a camera. Moreover, it is demonstrated that both NN-guided time-series encoding and rule-guided event detection contribute to overall classification accuracy. Full article
(This article belongs to the Special Issue Intelligent Vehicles: Overcoming Challenges)
►▼ Show Figures

Figure 1

Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-4
Back to TopTop