Deep Learning (DL) techniques are at the heart of most future advanced software functions in Critical Autonomous AI-based Systems (CAIS), where they also represent a major competitive factor. Hence, the economic success of CAIS industries (e.g., automotive, space, railway) depends on their ability to design, implement, qualify, and certify DL-based software products under bounded effort/cost. However, there is a fundamental gap between Functional Safety (FUSA) requirements on CAIS and the nature of DL solutions. This gap stems from the development process of DL libraries and affects high-level safety concepts such as (1) explainability and traceability, (2) suitability for varying safety requirements, (3) FUSA-compliant implementations, and (4) real-time constraints. As a matter of fact, the data-dependent and stochastic nature of DL algorithms clashes with current FUSA practice, which instead builds on deterministic, verifiable, and pass/fail test-based software. The SAFEXPLAIN project tackles these challenges and targets by providing a flexible approach to allow the certification - hence adoption - of DL-based solutions in CAIS building on: (1) DL solutions that provide end-to-end traceability, with specific approaches to explain whether predictions can be trusted and strategies to reach (and prove) correct operation, in accordance to certification standards; (2) alternative and increasingly sophisticated design safety patterns for DL with varying criticality and fault tolerance requirements; (3) DL library implementations that adhere to safety requirements; and (4) computing platform configurations, to regain determinism, and probabilistic timing analyses, to handle the remaining non-determinism.
This paper investigates the feasibility of using the periocular region for expression recognition. Most works have tried to solve this by analyzing the whole face. Periocular is the facial region in the immediate vicinity of the eye. It has the advantage of being available over a wide range of distances and under partial face occlusion, thus making it suitable for unconstrained or uncooperative scenarios. We evaluate five different image descriptors on a dataset of 1,574 images from 118 subjects. The experimental results show an average/overall accuracy of 67.0/78.0% by fusion of several descriptors. While this accuracy is still behind that attained with full-face methods, it is noteworthy to mention that our initial approach employs only one frame to predict the expression, in contraposition to state of the art, exploiting several order more data comprising spatial-temporal data which is often not available.
Platooning will soon likely to be common on Swedish roads and the potential for fuel savings in the transport sector is high. This pre-study project explores the need for external signaling in platoons to avoid any cut-ins from surrounding vehicles whose drivers are unaware that their actions may cause a loss of fuel saving. Interviews with truck drivers created an understanding of how they experience the behavior of the surrounding traffic. The scenarios that are highlighted where unaware cut-ins may occur are mainly on-ramps and while overtaking on highway. Car drivers highlighted that overtaking may be a problem, especially on 2+1 roads. Communication needs elicited in workshops with drivers mainly concerned the movement patterns and properties of vehicles, e.g. speed, direction, gaps and length of the platoon. Barriers that were identified for external signaling is that trailers are constantly rotating between different tractors. This may require that more trailers than tractors need to be equipped with communication devices. To evaluate the potential impact of external signaling simulation could be used, where a driving simulator could be used to evaluate the perception of car- and truck drivers. Different means of communication, behavior, driving close together or lighting could be subject to evaluation. The long-term learning effect and behavioral adaptation to platooning in traffic is also important to study. It was found that there are large regional behavioral differences in traffic. Naturalistic data from the US, indicate that there are no cut-ins if the distance between trucks is < 30 m. In Europe, the data collected from ETPC indicate that there is up to one cut-in every 15 km on highways. The data from the ETPC is however very sparse compared to the US study. In Sweden, it does not seem to be a specific need for external signaling since very few cut-ins occur. In Europe, more cut-ins occur and external signaling could help to save fuel. It is however unclear what long-term effects external signaling may have. Further studies are suggested to study if short platooning distance (10-20 meters) is sufficient to deter surrounding traffic from cut-ins.
For safety reasons, autonomous vehicles should communicate their intent rather than explicitly invitepeople to act. At RISE Viktoria in Sweden, we believe this simple design principle will impact howautonomous vehicles are experienced in the future
Platooning refers to a group of vehicles that--enabled by wireless vehicle-to-vehicle (V2V) communication and vehicle automation--drives with short inter-vehicular distances. Before its deployment on public roads, several challenging traffic situations need to be handled. Among the challenges are cut-in situations, where a conventional vehicle--a vehicle that has no automation or V2V communication--changes lane and ends up between vehicles in a platoon. This paper presents results from a simulation study of a scenario, where a conventional vehicle, approaching from an on-ramp, merges into a platoon of five cars on a highway. We created the scenario with four platooning gaps: 15, 22.5, 30, and 42.5 meters. During the study, the conventional vehicle was driven by 37 test persons, who experienced all the platooning gaps using a driving simulator. The participants' opinions towards safety, comfort, and ease of driving between the platoon in each gap setting were also collected through a questionnaire. The results suggest that a 15-meter gap prevents most participants from cutting in, while causing potentially dangerous maneuvers and collisions when cut-in occurs. A platooning gap of at least 30 meters yield positive opinions from the participants, and facilitating more smooth cut-in maneuvers while less collisions were observed.
Cut-in situations occurs when a vehicle intentionally changes lane and ends up in front of another vehicle or in-between two vehicles. In such situations, having a method to indicate the collision risk prior to making the cut-in maneuver could potentially reduce the number of sideswipe and rear end collisions caused by the cut-in maneuvers. This paper propose a new risk indicator, namely cut-in risk indicator (CRI), as a way to indicate and potentially foresee collision risks in cut-in situations. As an example use case, we applied CRI on data from a driving simulation experiment involving a manually driven vehicle and an automated platoon in a highway merging situation. We then compared the results with time-to-collision (TTC), and suggest that CRI could correctly indicate collision risks in a more effective way. CRI can be computed on all vehicles involved in the cut-in situations, not only for the vehicle that is cutting in. Making it possible for other vehicles to estimate the collision risk, for example if a cut-in from another vehicle occurs, the surrounding vehicles could be warned and have the possibility to react in order to potentially avoid or mitigate accidents.
This paper is an experience report of team Halmstad from the participation in a competition organised by the i-GAME project, the Grand Cooperative Driving Challenge 2016. The competition was held in Helmond, The Netherlands, during the last weekend of May 2016. We give an overview of our car's control and communication system that was developed for the competition following the requirements and specifications of the i-GAME project. In particular, we describe our implementation of cooperative adaptive cruise control, our solution to the communication and logging requirements, as well as the high level decision making support. For the actual competition we did not manage to completely reach all of the goals set out by the organizers as well as ourselves. However, this did not prevent us from outperforming the competition. Moreover, the competition allowed us to collect data for further evaluation of our solutions to cooperative driving. Thus, we discuss what we believe were the strong points of our system, and discuss post-competition evaluation of the developments that were not fully integrated into our system during competition time.
Cooperative adaptive cruise control (CACC) is a cooperative intelligent transport systems (C-ITS) function,which especially when used in platooning applications, possess many expected benefits including efficient road spaceutilization and reduced fuel consumption. Cut-in manoeuvres in platoons can potentially reduce those benefits, and are notdesired from a safety point of view. Unfortunately, in realistic traffic scenarios, cut-in manoeuvres can be expected, especiallyfrom non-connected vehicles. In this paper two different controllers for platooning are explored, aiming at maintaining thesafety of the platoon while a vehicle is cutting in from the adjacent lane. A realistic scenario, where a human driver performsthe cut-in manoeuvre is used to demonstrate the effectiveness of the controllers. Safety analysis of CACC controllers usingtime to collision (TTC) under such situation is presented. The analysis using TTC indicate that, although potential risks arealways high in CACC applications such as platooning due to the small inter-vehicular distances, dangerous TTC (TTC < 6seconds) is not frequent. Future research directions are also discussed along with the results.
It is currently unknown how automated vehicle platoons will be perceived by other road users in their vicinity. This study explores how drivers of manually operated passenger cars interact with automated passenger car platoons while merging onto a highway, and how different inter-vehicular gaps between the platooning vehicles affect their experience and safety. The study was conducted in a driving simulator and involved 16 drivers of manually operated cars. Our results show that the drivers found the interactions mentally demanding, unsafe, and uncomfortable. They commonly expected that the platoon would adapt its behavior to accommodate a smooth merge. They also expressed a need for additional information about the platoon to easier anticipate its behavior and avoid cutting-in. This was, however, affected by the gap size; larger gaps (30 and 42.5 m) yielded better experience, more frequent cut-ins, and less crashes than the shorter gaps (15 and 22.5 m). A conclusion is that a short gap as well as external human–machine interfaces (eHMI) might be used to communicate the platoon's intent to “stay together”, which in turn might prevent drivers from cutting-in. On the contrary, if the goal is to facilitate frequent, safe, and pleasant cut-ins, gaps larger than 22.5 m may be suitable. To thoroughly inform such design trade-offs, we urge for more research on this topic. © 2021 The Author(s)
In the near future, Cooperative Intelligent Transport System (C-ITS) applications are expected to be deployed. To support this, simulation is often used to design and evaluate the applications during the early development phases. Simulations of C-ITS scenarios often assume a fleet of homogeneous vehicles within the transportation system. In contrast, once C-ITS is deployed, the traffic scenarios will consist of a mixture of connected and non-connected vehicles, which, in addition, can be driven manually or automatically. Such mixed cases are rarely analysed, especially those where manually driven vehicles are involved. Therefore, this paper presents a C-ITS simulation framework, which incorporates a manually driven car through a driving simulator interacting with a traffic simulator, and a communication simulator, which together enable modelling and analysis of C-ITS applications and scenarios. Furthermore, example usages in the scenarios, where a manually driven vehicle cut-in to a platoon of Cooperative Adaptive Cruise Control (CACC) equipped vehicles are presented.
Millions of vehicles are transported every year, tightly parked in vessels or boats. To reduce the risks of associated safety issues like fires, knowing the location of vehicles is essential, since different vehicles may need different mitigation measures, e.g. electric cars. This work is aimed at creating a solution based on a nano-drone that navigates across rows of parked vehicles and detects their license plates. We do so via a wall-following algorithm, and a CNN trained to detect license plates. All computations are done in real-time on the drone, which just sends position and detected images that allow the creation of a 2D map with the position of the plates. Our solution is capable of reading all plates across eight test cases (with several rows of plates, different drone speeds, or low light) by aggregation of measurements across several drone journeys.
With the emerging connected automated vehicles, 5G and Internet of Things (IoT), vehicles and road infrastructure become connected and cooperative, enabling Cooperative Intelligent Transport Systems (C-ITS). C-ITS are transport system of systems that involves many stakeholders from different sectors. While running their own systems and providing services independently, stakeholders cooperate with each other for improving the overall transport performance such as safety, efficiency and sustainability. Massive information on road and traffic is already available and provided through standard services with different protocols. By reusing and composing the available heterogeneous services, novel value-added applications can be developed. This paper introduces a choreography-based service composition platform, i.e. the CHOReVOLUTION Integrated Development and Runtime Environment (IDRE), and it reports on how the IDRE has been successfully exploited to accelerate the reuse-based development of a choreography-based Urban Traffic Coordination (UTC) application. The UTC application takes the shape of eco-driving services that through real-time eco-route evaluation assist the drivers for the most eco-friendly and comfortable driving experience. The eco-driving services are realized through choreography and they are exploited through a mobile app for online navigation. From specification to deployment to execution, the CHOReVOLUTION IDRE has been exploited to support the realization of the UTC application by automatizing the generation of the distributed logic to properly bind, coordinate and adapt the interactions of the involved parties. The benefits brought by CHOReVOLUTION IDRE have been assessed through the evaluation of a set of Key Performance Indicators (KPIs).
An interaction protocol for cooperative platoon merge on highways is proposed. The interaction protocol facilitates a challenge scenario for the Grand Cooperative Driving Challenge (GCDC) 2016, where two platoons running on separate lanes merge into one platoon due to a roadwork in one of the lanes. Detailed interaction procedures, described with state machines of each vehicle are presented. A communication message set is designed to support platoon controllers to perform safe and efficient manoeuvres.
This paper presents an overview of current projects that deal with vehicle platooning. The platooning concept can be defined as a collection of vehicles that travel together, actively coordinated in formation. Some expected advantages of platooning include increased fuel and traffic efficiency, safety and driver comfort. There are many variations of the details of the concept such as: the goals of platooning, how it is implemented, mix of vehicles, the requirements on infrastructure, what is automated (longitudinal and lateral control) and to what level. The following projects are presented: SARTRE â a European platooning project; PATH â a California traffic automation program that includes platooning; GCDC â a cooperative driving initiative, SCANIA platooning and; Energy ITS â a Japanese truck platooning project.
Deep neural networks (DNNs) will emerge as a cornerstone in automotive software engineering. However, developing systems with DNNs introduces novel challenges for safety assessments. This paper reviews the state-of-the-art in verification and validation of safety-critical systems that rely on machine learning. Furthermore, we report from a workshop series on DNNs for perception with automotive experts in Sweden, confirming that ISO 26262 largely contravenes the nature of DNNs. We recommend aerospace-to-automotive knowledge transfer and systems-based safety approaches, for example, safety cage architectures and simulated system test cases.
With the fifth generation (5G) communication technologies on the horizon, the society is rapidly transformed into a fully connected world. The Future Internet (FI) is foreseeable to consist of an infinite number of software components and things that coordinate with each other to enable different applications. Transport systems, as one of the most important systems in future smart cities, will embrace the connectivity, together with the fast development of cooperative and automated vehicles to enable smart traffic. To facilitate this transformation, a service choreography composition platform is under development to enable fast innovation and prototyping of choreography-based Internet of Things (IoT) applications by automatically synthesizing choreographies. Based on the method, a smart traffic application is developed and demonstrated.
Intersection management is one of the most challenging problems within the transport system. Traffic light-based methods have been efficient but are not able to deal with the growing mobility and social challenges. On the other hand, the advancements of automation and communications have enabled cooperative intersection management, where road users, infrastructure, and traffic control centers are able to communicate and coordinate the traffic safely and efficiently. Major techniques and solutions for cooperative intersections are surveyed in this paper for both signalized and nonsignalized intersections, whereas focuses are put on the latter. Cooperative methods, including time slots and space reservation, trajectory planning, and virtual traffic lights, are discussed in detail. Vehicle collision warning and avoidance methods are discussed to deal with uncertainties. Concerning vulnerable road users, pedestrian collision avoidance methods are discussed. In addition, an introduction to major projects related to cooperative intersection management is presented. A further discussion of the presented works is given with highlights of future research topics. This paper serves as a comprehensive survey of the field, aiming at stimulating new methods and accelerating the advancement of automated and cooperative intersections.
With intensive research and field operational tests over the intelligent transportation area and the advancements of information and communication technologies, intelligent transportation systems reach the stage of deployment. EU focuses on cooperative intelligent transportation systems and confirms the finalization of the first release of the standards, paving the way for deployment in the coming years. This paper presents the concept of EU's cooperative intelligent transportation systems and describes in detail the functional architecture, together with highlights of related standardsthat have been finalized in Release 1. Latest updates of the cooperative intelligent transportation systems are provided for both industry and academia, aiming at helping to accelerate cooperative mobility.
Emergency management has long been recognized as a social challenge due to the criticality of the response time. In emergency situations such as severe traffic accidents, minimizing the response time, which requires close collaborations between all stakeholders involved and distributed intelligence support, leads to greater survival chance of the injured. However, the current response system is far from efficient, despite the rapid development of information and communication technologies. This paper presents an automated collaboration framework for emergency management that coordinates all stakeholders within the emergency response system and fully automates the rescue process. Applying the concept of multiaccess edge computing architecture, as well as choreography of the service oriented architecture, the system allows seamless coordination between multiple organizations in a distributed way through standard web services. A service choreography is designed to globally model the emergency management process from the time an accident occurs until the rescue is finished. The choreography can be synthesized to generate detailed specification on peer-to-peer interaction logic, and then the specification can be enacted and deployed on cloud infrastructures.
Despite the existing regulation efforts and measures, vehicles with dangerous goods still pose significant risks on public safety, especially in road tunnels. Solutions based on cooperative intelligent transportation system (C-ITS) are promising measures, however, they have received limited attention. We propose C-ITS applications that coordinate dangerous goods vehicles to minimize the risk by maintaining safe distances between them in road tunnels. Different mechanisms, including global centralized coordination, global distributed coordination, and local coordination, are proposed and investigated. A preliminary simulation is performed and demonstrates their effectiveness.
Despite the strong interests in creating data economy, automotive industries are creating data silos with each stakeholder maintaining its own data cloud. Federated learning (FL), designed for privacy-preserving collaborative Machine Learning (ML), offers a promising method that allows multiple stakeholders to share information through ML models without the exposure of raw data, thus natively protecting privacy. Motivated by the strong need for automotive collaboration and the advancement of FL, this paper investigates how FL could enable privacy-preserving information sharing for automotive industries. We first introduce the statuses and challenges for automotive data sharing, followed by a brief introduction to FL. We then present a comprehensive discussion on potential applications of federated learning to enable an automotive collaborative ecosystem. To illustrate the benefits, we apply FL for driver action classification and demonstrate the potential for collaborative machine learning without data sharing.
Modeling the interaction of vehicles during certain traffic situations is the starting point for creating autonomous driving. Data collected from field trials where test subjects drive through a single-vehicle intersection was used to create behavioral models. The present work describes two implementations of models based on the dynamical systems approach and compares similarities and differences between them. The proposed models are designed to closely replicate the behavior selection in the intersection crossing experiment.
Action intention recognition is becoming increasingly important in the road vehicle automation domain. Autonomous vehicles must be aware of their surroundings if we are to build safe and efficient transport systems. This paper explores methods for predicting the action intentions of road users based on an aware and intelligent 3D camera-based sensor system. The collected data contains trajectories of two different scenarios. The first one includes bicyclists and the second cars that are driving in a road approaching an intersection where they are either turning or continuing straight. The data acquisition system is used to collect trajectories of the road users that are used as input for models trained to predict the action intention of the road users.
Smart cities and communities (SCC) constitute a new paradigm in urban development. SCC ideate a data-centered society aimed at improving efficiency by automating and optimizing activities and utilities. Information and communication technology along with Internet of Things enables data collection and with the help of artificial intelligence (AI) situation awareness can be obtained to feed the SCC actors with enriched knowledge. This paper describes AI perspectives in SCC and gives an overview of AI-based technologies used in traffic to enable road vehicle automation and smart traffic control. Perception, smart traffic control and driver modeling are described along with open research challenges and standardization to help introduce advanced driver assistance systems and automated vehicle functionality in traffic. To fully realize the potential of SCC, to create a holistic view on a city level, availability of data from different stakeholders is necessary. Further, though AI technologies provide accurate predictions and classifications, there is an ambiguity regarding the correctness of their outputs. This can make it difficult for the human operator to trust the system. Today there are no methods that can be used to match function requirements with the level of detail in data annotation in order to train an accurate model. Another challenge related to trust is explainability: models can have difficulty explaining how they came to certain conclusions, so it is difficult for humans to trust them.
The Grand Cooperative Driving Challenge (GCDC), with the aim to boost the introduction of cooperative automated vehicles by means of wireless communication, is presented. Experiences from the previous edition of GCDC, which was held in Helmond in the Netherlands in 2011, are summarized, and an overview and expectations of the challenges in the 2016 edition are discussed. Two challenge scenarios, cooperative platoon merge and cooperative intersection passing, are specified and presented. One demonstration scenario for emergency vehicles is designed to showcase the benefits of cooperative driving. Communications closely follow the newly published cooperative intelligent transport system standards, while interaction protocols are designed for each of the scenarios. For the purpose of interoperability testing, an interactive testing tool is designed and presented. A general summary of the requirements on teams for participating in the challenge is also presented.
Current transportation systems face great challenges due to the increasing mobility. Traffic accidents, congestion, air pollution, etc., are all calling for new methods to improve the transportation system. With the US legislation in progress over vehicle communications and EU’s finalization of the basic set of standards over cooperative intelligent transportation systems (C-ITS), vehicular ad hoc network (VANET) based applications are expected to address those challenges and provide solutions for a safer, more efficient and sustainable future intelligent transportation systems (ITS). In this chapter, transportation challenges are firstly summarized in respect of safety, efficiency, environmental threat, etc. A brief introduction of the VANET is discussed along with state of the art of VANET-based applications. Based on the current progress and the development trend of VANET, a number of new features of future VANET are identified, together with a set of potential future ITS applications. The on-going research and field operational test projects, which are the major enabling efforts for the future VANET-based C-ITS, are presented. The chapter is of great interest to readers working within ITS for current development status and future trend within the C-ITS area. It is also of interest to general public for an overview of the VANET enabled future transportation system.
Cooperative speed harmonization based on floating car data aiming at improving manoeuvrability in a highly utilized intersection is presented. Cooperative Intelligent Transportation Systems (C-ITS) aims at gather information about the current traffic situation based on wireless communication and provide aggregated information back to the road users in order to improve e.g. efficiency, safety and/or comfort. Simulations show that the proposed speed harmonization application is capable of lowering the CO2 emissions with up to 11%, increasing the average speed with up to 14% and reducing the travel time with up to 16% for all vehicles in the simulation. It is also found that not only the cooperative vehicles benefit from the application but also the non-equipped vehicles. Furthermore, the cooperative traffic simulator has been shown to be a valuable tool for investigating how C-ITS applications may be utilized to develop future traffic system.
Free-floating car sharing is a form of car rental used by people for short periods of time where the cars can be picked up and returned anywhere within a given area. In this paper, we have collected free-floating car sharing data, for electric as well as fossil fueled cars, and data regarding e.g. size of the city, number of cars in the service, etc. The utilization rates of the free-floating car sharing services vary much between the cities, greatly influencing the success of the services. This paper presents the most important factors influencing the utilization rate, and also a methodology to predict the utilization rate for new cities, using data mining based on Random Forests.
Technology is to a large extent driving the development of road vehicle automation. This Chapter summarizes the general overall trends in the enabling technologies within this field that were discussed during the Enabling technologies for road vehicle automation breakout session at the Automated Vehicle Symposium 2016. With a starting point in six scenarios that have the potential to be deployed at an early stage, five different categories of emerging technologies are described: (a) positioning, localization and mapping (b) algorithms, deep learning techniques, sensor fusion guidance and control (c) hybrid communication (d) sensing and perception and (e) technologies for data ownership and privacy. It is found that reliability and extensive computational power are the two most common challenges within the emerging technologies. Furthermore, cybersecurity binds all technologies together as vehicles will be constantly connected. Connectivity allows both improved local awareness through vehicle-to-vehicle communication and it allows continuous deployment of new software and algorithms that constantly learns new unforeseen objects or scenarios. Finally, while five categories were individually considered, further holistic work to combine them in a systems concept would be the important next step toward implementation.
Data mining techniques based on Random forests are explored to gain knowledge about data in a Field Operational Test (FOT) database. We compare the performance of a Random forest, a Support Vector Machine and a Neural network used to separate drowsy from alert drivers. 25 variables from the FOT data was utilized to train the models. It is experimentally shown that the Random forest outperforms the other methods while separating drowsy from alert drivers. It is also shown how the Random forest can be used for variable selection to find a subset of the variables that improves the classification accuracy. Furthermore it is shown that the data proximity matrix estimated from the Random forest trained using these variables can be used to improve both classification accuracy, outlier detection and data visualization.
This paper presents a novel approach to modelling visual distraction of bicyclists. A unique bicycle simulator equipped with sensors capable of capturing the behaviour of the bicyclist is presented. While cycling two similar scenario routes, once while simultaneously interacting with an electronic device and once without any electronic device, statistics of the measured speed, head movements, steering angle and bicycle road position along with questionnaire data are captured. These variables are used to model the self-assessed distraction level of the bicyclist. Data mining techniques based on random forests, support vector machines and neural networks are evaluated for the modelling task. Out of the total 71 measured variables a variable selection procedure based on random forests is able to select a fraction of those and consequently improving the modelling performance. By combining the random forest-based variable selection and support vector machine-based modelling technique the best overall performance is achieved. The method shows that with a few observable variables it is possible to use machine learning to model, and thus predict, the distraction level of a bicyclist.
A data proximity matrix is an important information source in random forests (RF) based data mining, including data clustering, visualization, outlier detection, substitution of missing values, and finding mislabeled data samples. A novel approach to estimate proximity is proposed in this work. The approach is based on measuring distance between two terminal nodes in a decision tree. To assess the consistency (quality) of data proximity estimate, we suggest using the proximity matrix as a kernel matrix in a support vector machine (SVM), under the assumption that a matrix of higher quality leads to higher classification accuracy. It is experimentally shown that the proposed approach improves the proximity estimate, especially when RF is made of a small number of trees. It is also demonstrated that, for some tasks, an SVM exploiting the suggested proximity matrix based kernel, outperforms an SVM based on a standard radial basis function kernel and the standard proximity matrix based kernel.
In recent years, free-floating car sharing services (FFCS) have been offered by many organizations as a moreflexible option compared to traditional car sharing. FFCS allows users to pick up and return cars anywherewithin a specified area of a city. FFCS can provide a high degree of utilization of vehicles and less usage ofinfrastructure in the form of parking lots and roads and thus has the potential to increase the efficiency of thetransport sector. However, there is also a concern that these compete with other efficient modes of transport suchas biking and public transport. The aim of this paper is to better understand how, when and where the vehiclesare utilized through logged data of the vehicles movements. We have access to data collected on FFCS servicesin 22 cities in Europe and North America which allows us to compare the usage pattern in different cities andexamine whether or not there are similar trends. In this paper, we use the collected data to compare the differentcities based on utilization rate, length of trip and time of day that the trip is made. We find that the vehicleutilization rates differ between cities with Madrid and Hamburg having some of the highest utilization levels forthe FFCS vehicles. The result form a first step of a better understanding on how these services are being usedand can provide valuable input to local policy makers as well as future studies such as simulation models.
Free-floating car sharing services (FFCS) have been offered as a more flexible mobility solution than other car sharing services. FFCS users can pick up and return cars anywhere within a specified area in a city.The objective of this paper is to identify similar usage patterns of FFCS in different cities as well as city characteristics that make these services a viable option. The authors have access to real booking data for 32 cities in Europe and North America. Their study shows the share of daily car trips is negatively correlated to the utilization rate of these services. Also, the higher the congestion and the harder finding a parking lot, the lower the utilization rate of these services is in the cities. Moreover, our results suggest that FFCS services do not compete with public transport but are rather used in combination to it. These services are mainly used during midday and evening peak and the trips taken by these services are mainly chained trips.The clustering analysis shows that the trips are grouped into two or three clusters in different cities. The majority of clusters are the inner city clusters which contain a significantly higher number of trips than the clusters around other points of interest such as airports.
How to ensure trust and societal acceptance of automated vehicles (AVs) is a widely-discussed topic today. While trust and acceptance could be influenced by a range of factors, one thing is sure: the ability of AVs to safely and smoothly interact with other road users will play a key role. Based on our experiences from a series of studies, this paper elaborates on issues that AVs may face in interactions with other road users and whether external vehicle interfaces could support these interactions. Our overall conclusion is that such interfaces may be beneficial in situations where negotiation is needed. However, these benefits, and potential drawbacks, need to be further explored to create a common language, or standard, for how AVs should communicate with other road users.
The aim of this study is to identify current research gaps, challenges, and opportunities in the field of vehicle automation. The study is based on a literature review. The review shows that the current research focuses mainly on improvements in sensing, actuation, and navigation systems. However, this study acknowledges a range of challenges in other areas that need to be addressed to facilitate possible benefits that vehicle automation may bring. In particular, the following challenges are highlighted: 1) understanding the transfer of control between the vehicle and the driver, and vice versa, 2) defining behavior of automated vehicles in relation to other road users (e.g., pedestrians), 3) identifying how to communicate the system reliability information to the drivers, and 4) clarifying the impact on societal values, i.e. what driver behaviors that will be considered as appropriate, or even acceptable. The work presented here is a part of the ongoing project Boundary Conditions for Vehicle Automation, co-financed by the Swedish Governmental Agency for Innovation Systems (VINNOVA) and carried out by SAFER-Vehicle and Traffic Safety Centre at Chalmers.
Several areas have been improved with Deep Learning during the past years. For non-safety related products adoption of AI and ML is not an issue, whereas in safety critical applications, robustness of such approaches is still an issue. A common challenge for Deep Neural Networks (DNN) occur when exposed to out-of-distribution samples that are previously unseen, where DNNs can yield high confidence predictions despite no prior knowledge of the input. In this paper we analyse two supervisors on two well-known DNNs with varied setups of training and find that the outlier detection performance improves with the quality of the training procedure. We analyse the performance of the supervisor after each epoch during the training cycle, to investigate supervisor performance as the accuracy converges. Understanding the relationship between training results and supervisor performance is valuable to improve robustness of the model and indicates where more work has to be done to create generalized models for safety critical applications.
Deep Neural Networks (DNN) have improved the quality of several non-safety related products in the past years. However, before DNNs should be deployed to safety-critical applications, their robustness needs to be systematically analyzed. A common challenge for DNNs occurs when input is dissimilar to the training set, which might lead to high confidence predictions despite proper knowledge of the input. Several previous studies have proposed to complement DNNs with a supervisor that detects when inputs are outside the scope of the network. Most of these supervisors, however, are developed and tested for a selected scenario using a specific performance metric. In this work, we emphasize the need to assess and compare the performance of supervisors in a structured way. We present a framework constituted by four datasets organized in six test cases combined with seven evaluation metrics. The test cases provide varying complexity and include data from publicly available sources as well as a novel dataset consisting of images from simulated driving scenarios. The latter we plan to make publicly available. Our framework can be used to support DNN supervisor evaluation, which in turn could be used to motive development, validation, and deployment of DNNs in safety-critical applications.
Context: Deep Neural Networks (DNN) have shown great promise in various domains, for example to support pattern recognition in medical imagery. However, DNNs need to be tested for robustness before being deployed in safety critical applications. One common challenge occurs when the model is exposed to data samples outside of the training data domain, which can yield to outputs with high confidence despite no prior knowledge of the given input. Objective: The aim of this paper is to investigate how the performance of detecting out-of-distribution (OOD) samples changes for outlier detection methods (e.g., supervisors) when DNNs become better on training samples. Method: Supervisors are components aiming at detecting out-of-distribution samples for a DNN. The experimental setup in this work compares the performance of supervisors using metrics and datasets that reflect the most common setups in related works. Four different DNNs with three different supervisors are compared during different stages of training, to detect at what point during training the performance of the supervisors begins to deteriorate. Results: Found that the outlier detection performance of the supervisors increased as the accuracy of the underlying DNN improved. However, all supervisors showed a large variation in performance, even for variations of network parameters that marginally changed the model accuracy. The results showed that understanding the relationship between training results and supervisor performance is crucial to improve a model's robustness. Conclusion: Analyzing DNNs for robustness is a challenging task. Results showed that variations in model parameters that have small variations on model predictions can have a large impact on the out-of-distribution detection performance. This kind of behavior needs to be addressed when DNNs are part of a safety critical application and hence, the necessary safety argumentation for such systems need be structured accordingly.
Machine learning (ML) applications generate a continuous stream of success stories from various domains. ML enables many novel applications, also in safety-critical contexts. However, the functional safety standards such as ISO 26262 did not evolve to cover ML. We conduct an exploratory study on which parts of ISO 26262 represent the most critical gaps between safety engineering and ML development. While this paper only reports the first steps toward a larger research endeavor, we report three adaptations that are critically needed to allow ISO 26262 compliant engineering, and related suggestions on how to evolve the standard.
In this paper, we present the Cooperative Adaptive Cruise Control (CACC) architecture, which was proposed and im- plemented by the team from Chalmers University of Technology, Göteborg, Sweden, that joined the Grand Cooperative Driving Challenge (GCDC) in 2011. The proposedCACCarchitecture con- sists of the following three main components, which are described in detail: 1) communication; 2) sensor fusion; and 3) control. Both simulation and experimental results are provided, demonstrating that the proposed CACC system can drive within a vehicle platoon while minimizing the inter-vehicle spacing within the allowed range of safety distances, tracking a desired speed profile, and attenuating acceleration shockwaves.