Deep Neural Network Perception Models and Robust Autonomous Driving Systems (original) (raw)
Related papers
Current challenges in autonomous driving
IOP Conference Series: Materials Science and Engineering, 2017
Nowadays the automotive industry makes a quantum shift to a future, where the driver will have smaller and smaller role in driving his or her vehicle ending up being totally excluded. In this paper, we have investigated the different levels of driving automatization, the prospective effects of these new technologies on the environment and traffic safety, the importance of regulations and their current state, the moral aspects of introducing these technologies and the possible scenarios of deploying the autonomous vehicles. We have found that the self-driving technologies are facing many challenges: a) They must make decisions faster in very diverse conditions which can include many moral dilemmas as well; b) They have an important potential in reducing the environmental pollution by optimizing their routes, driving styles by communicating with other vehicles, infrastructures and their environment; c) There is a considerable gap between the self-drive technology level and the current regulations; fortunately, this gap shows a continuously decreasing trend; d) In case of many types of imminent accidents management there are many concerns about the ability of making the right decision. Considering that this field has an extraordinary speed of development, our study is up to date at the submission deadline. Self-driving technologies become increasingly sophisticated and technically accessible, and in some cases, they can be deployed for commercial vehicles as well. According to the current stage of research and development, it is still unclear how the self-driving technologies will be able to handle extreme and unexpected events including their moral aspects. Since most of the traffic accidents are caused by human error or omission, it is expected that the emergence of the autonomous technologies will reduce these accidents in their number and gravity, but the very few currently available test results have not been able to scientifically underpin this issue yet. The increasing trend in automation of vehicles will radically change the composition of car industry players, as mechatronics will not only be a complementary part of the automobile industry but an indispensable part of it. There is a reasonable expectation that automated cars will perform the same or better in all respects than their conventional counterparts. However, it seems that the current regulations do not keep up with the development of technology and sometimes hinder the development and testing of autonomous technologies.
Towards Fully Autonomous Driving ? The Perception & Decision-making bottleneck
2016
Thanks to the DARPA Urban Challenge in 2007, the launch a few years later of the Google Car project, the recent multiple announcements of the Automotive Industry, and the wide dissemination by the media of nice interviews and videos, the concept of Autonomous Driving seems to progressively become a reality for the next decades. The objective of this talk is to give a brief analysis of the state of the art, before focusing on one of the current brake on the deployment of such a technology: the lack of robustness and of efficiency of current Embedded Perception and Decision-making systems. New technologies and trends will be presented for addressing these issues. It will be shown that Bayesian approaches are mandatory for developing such technologies and for obtaining the required robustness in presence of uncertainty and complex dynamic situations. The talk will be illustrated by some interesting results obtained at Inria in the scope of several collaborative projects involving Toyot...
Autonomous Driving — a Challenge for the Automotive Industry
Intereconomics
ZBW-Leibniz Information Centre for Economics 171 Automotive Industry ers and the position of suppliers competing for a share of the autonomous vehicles market. 2 Autonomous vehicles Autonomous driving takes a wide variety of forms. The classifi cation produced by the Society of Automotive Engineers (Table 1) has attracted considerable attention. 3 The German Federal Highway Research Institute devised a similar taxonomy that differentiates vehicles driven solely by the driver (manual driving), driver-assisted, semiautomated, highly automated and fully automated. 4 The precise differences in the defi nitions are not discussed here in greater detail, but the decisive factor is the unanimous opinion that there is no clear-cut, binary distinction between automated and non-automated vehicles. Rather, both describe a continuum of progressive steps of automation. Simply using a lane departure warning system, for example, is not considered automation. Lane departure assis-2 This paper updates H. B a r d t : Autonomes Fahren-Eine Herausforderung für die deutsche Autoindustrie, in: IW-Trends-Vierteljahresschrift zur empirischen Wirtschaftsforschung aus dem Institut der
Autonomous Cars: Developments, Technical Challenges and Opportunities
International Journal for Research in Applied Science & Engineering Technology, 2021
Machine Language is being used in technological advancement in all the domains nowadays, automotive domain being one of them. But practically implementing the ML has its own challenges in the automotive domain since it concerns the lives of people in the car, pedestrians and other vehicles. Even though AI had been in existence for long, the challenges are so high that still we are not in a stage where Level 5 autonomous cars can be deployed in any part of the globe. Here we look into the various challenges faced in the Autonomous cars and different implementations performed.
A Survey of Autonomous Driving: Common Practices and Emerging Technologies
IEEE Access, 2020
Automated driving systems (ADSs) promise a safe, comfortable and efficient driving experience. However, fatalities involving vehicles equipped with ADSs are on the rise. The full potential of ADSs cannot be realized unless the robustness of state-of-the-art is improved further. This paper discusses unsolved problems and surveys the technical aspect of automated driving. Studies regarding present challenges, high-level system architectures, emerging methodologies and core functions including localization, mapping, perception, planning, and human machine interfaces, were thoroughly reviewed. Furthermore, many stateof-the-art algorithms were implemented and compared on our own platform in a real-world driving setting. The paper concludes with an overview of available datasets and tools for ADS development.
Use of CNN(YOLO) in Self Driving vehicles
The field of autonomous automation is of interest to researchers, and much has been accomplished in this area, of which this paper presents a detailed chronology. This paper can help one understand the trends in autonomous vehicle technology for the past, present, and future. We see a drastic change in autonomous vehicle technology since 1920s, when the first radio controlled vehicles were designed. In the subsequent decades, we see fairly autonomous electric cars powered by embedded circuits in the roads. By 1960s, autonomous cars having similar electronic guide systems came into picture. 1980s saw vision guided autonomous vehicles, which was a major milestone in technology and till date we use similar or modified forms of vision and radio guided technologies. Various semi-autonomous features introduced in modern cars such as lane keeping, automatic braking and adaptive cruise control are based on such systems. Extensive network guided systems in conjunction with vision guided features is the future of autonomous vehicles. It is predicted that most companies will launch fully autonomous vehicles by the advent of next decade. The future of autonomous vehicles is an ambitious era of safe and comfortable transportation. Background An autonomous car is a vehicle capable of sensing its environment and operating without human involvement. A human passenger is not required to take control of the vehicle at any time, nor is a human passenger required to be present in the vehicle at all. An autonomous car can go anywhere a traditional car goes and do everything that an experienced human driver does.The Society of Automotive Engineers (SAE) currently defines 6 levels of driving automation ranging from Level 0 (fully manual) to Level 5 (fully autonomous). These levels have been adopted by the U.S. Department of Transportation.
The Ecosystem of the Next-Generation Autonomous Vehicles
Advances in Science, Technology and Engineering Systems Journal, 2021
Autonomous vehicles (AVs) technology is expected to provide many benefits for the society such as providing safe transportation to the community and reducing the number of accidents on the roads. With the emergence of AVs, the conventional safety infrastructure in which humans drive their vehicles will need to be upgraded in order to take full advantage of the new technology. AVs are responsible for the different aspects of driving, namely: perception, decision making and taking action. Capturing data and diagnosing issues becomes imperative in this new transportation system because an error might be an indication of a systemic problem which may lead to future accidents or failures. Therefore, any AV accident must be dealt with seriously. Unfortunately, current procedures and the type of data collected during the investigation of an accident is not sufficient; For example, AV minor accidents are not investigated in depth as it is with AV major accidents and therefore, if the cause of the accident is a systemic issue, then it might cause more accidents in the future. The main goal of this paper is to explore the requirements for accident reports for incidents involving AVs and the procedures to escalate issues to avoid systemic risks. All the information available about AV accidents, regardless of the severity of the accident, will be analyzed and studied. This paper will present three recommendations; An updated law enforcement accident report, an escalation procedure that depends on the diagnosis of the fault and a database of AV accidents to enable ongoing learning to find systemic issues.
2020
Technological development within the domain of autonomous driving systems remains an important area of focus for government, industry, and academia. Large corporations are continuing research and development in unmanned delivery platforms; this technology holds the potential to substantially decrease costs and increase efficiency in supply chain management. For the military, autonomous systems show the potential to increase capability and improve Soldier safety. The West Point Autonomous Vehicle Research and Design (AVRAD) team developed an autonomous system, capable of competing in the 2020 Intelligent Ground Vehicle Competition (IGVC)--Self Drive Competition. IGVC leadership decided to cancel the 202 IGVC due to the COVID-19 pandemic. However, for the past 27 years, this international competition challenged teams to think creatively about evolving technologies of vehicle sensors, robotics, and system integration. During the event, teams race their autonomous systems through a cour...
Autonomous vehicles: from paradigms to technology
IOP Conference Series: Materials Science and Engineering, 2017
Mobility is a basic necessity of contemporary society and it is a key factor in global economic development. The basic requirements for the transport of people and goods are: safety and duration of travel, but also a number of additional criteria are very important: energy saving, pollution, passenger comfort. Due to advances in hardware and software, automation has penetrated massively in transport systems both on infrastructure and on vehicles, but man is still the key element in vehicle driving. However, the classic concept of 'human-in-the-loop' in terms of 'hands on' in driving the cars is competing aside from the self-driving startups working towards so-called 'Level 4 autonomy', which is defined as "a self-driving system that does not requires human intervention in most scenarios". In this paper, a conceptual synthesis of the autonomous vehicle issue is made in connection with the artificial intelligence paradigm. It presents a classification of the tasks that take place during the driving of the vehicle and its modeling from the perspective of traditional control engineering and artificial intelligence. The issue of autonomous vehicle management is addressed on three levels: navigation, movement in traffic, respectively effective maneuver and vehicle dynamics control. Each level is then described in terms of specific tasks, such as: route selection, planning and reconfiguration, recognition of traffic signs and reaction to signaling and traffic events, as well as control of effective speed, distance and direction. The approach will lead to a better understanding of the way technology is moving when talking about autonomous cars, smart/intelligent cars or intelligent transport systems.
Deep Learning: Is it the Main Challenge Behind Autonomous Vehicles Deployment?
IEEE Technology Policy and Ethics
There has been a growing interest in the field of intelligent transportation systems (ITSs) to improve road safety [1] and traffic management issues. ITS is realized through social interaction among vehicles which offers plethora of applications and services ranging from safety to information and entertainment (collectively referred to as infotainment). These applications on one hand guarantee safe driving, and on the other hand add value to our driving experience. Significant efforts have been made in the last decade by researchers from both academia and industry to realize the ITS through different communication paradigms such as vehicle to vehicle (V2V) and vehicle to infrastructure (V2I) communications. ITS uses existing communication technologies such as Dedicated Short-Range Communication (DSRC), WiFi, 4G/LTE, Bluetooth, WiMax, and so on.