The last decade witnessed increasingly rapid progress in self‐driving vehicle technology, mainly backed up by advances in the area of deep learning and artificial intelligence (AI). The objective of ...this paper is to survey the current state‐of‐the‐art on deep learning technologies used in autonomous driving. We start by presenting AI‐based self‐driving architectures, convolutional and recurrent neural networks, as well as the deep reinforcement learning paradigm. These methodologies form a base for the surveyed driving scene perception, path planning, behavior arbitration, and motion control algorithms. We investigate both the modular perception‐planning‐action pipeline, where each module is built using deep learning methods, as well as End2End systems, which directly map sensory information to steering commands. Additionally, we tackle current challenges encountered in designing AI architectures for autonomous driving, such as their safety, training data sources, and computational hardware. The comparison presented in this survey helps gain insight into the strengths and limitations of deep learning and AI approaches for autonomous driving and assist with design choices.
Self‐driving vehicles (SDVs) promise to considerably reduce traffic crashes. One pressing concern facing the public, automakers, and governments is “How safe is safe enough for SDVs?” To answer this ...question, a new expressed‐preference approach was proposed for the first time to determine the socially acceptable risk of SDVs. In our between‐subject survey (N = 499), we determined the respondents’ risk‐acceptance rate of scenarios with varying traffic‐risk frequencies to examine the logarithmic relationships between the traffic‐risk frequency and risk‐acceptance rate. Logarithmic regression models of SDVs were compared to those of human‐driven vehicles (HDVs); the results showed that SDVs were required to be safer than HDVs. Given the same traffic‐risk‐acceptance rates for SDVs and HDVs, their associated acceptable risk frequencies of SDVs and HDVs were predicted and compared. Two risk‐acceptance criteria emerged: the tolerable risk criterion, which indicates that SDVs should be four to five times as safe as HDVs, and the broadly acceptable risk criterion, which suggests that half of the respondents hoped that the traffic risk of SDVs would be two orders of magnitude lower than the current estimated traffic risk. The approach and these results could provide insights for government regulatory authorities for establishing clear safety requirements for SDVs.
Self-driving vehicles are of critical importance to a future sustainable transport system, which is expected to become widespread around the world. However a substantial amount of risk is associated ...with self-driving vehicles which must be considered by decision-makers effectively. Given that automated driving technology and how it will interact with the mobility system are substantially risky, the risks involved in self-driving vehicles need to be addressed appropriately. The identified knowledge gap of the pre-literature review is that an overview of the identification which completely considers all types of risks related to self-driving vehicles does not exist. In response to this knowledge gap, this study aims to prioritize the risks in self-driving vehicles. Risk prioritization is a complicated multi-criteria decision making (MCDM) problem that requires consideration of multiple feasible alternatives and conflicting tangible and intangible criteria. This study addresses the prioritization of risks involved with self-driving vehicles by proposing new hybrid MCDM methods based on the Analytic Hierarchy Process (AHP), the Technique for order preference by similarity to an ideal solution (TOPSIS) and Vlse Kriterijumska Optimizacija I Kompromisno Resenje (VIKOR) under Pythagorean fuzzy environment. The result of the proposed model is validated by performing sensitivity analysis. The performance of proposed methodology with Pythagorean fuzzy sets is also compared with those with ordinary fuzzy sets and it is revealed that the proposed method produces reliable and informative outcomes better representing the impreciseness of decision making problems. The findings of this study will provide useful insight to the planners and policymakers for decision making in self-driving vehicles.
Display omitted
•This study addresses the prioritization of risks in self-driving vehicles.•A real-world multiple criteria decision making problem was introduced.•Hybrid approaches based on Pythagorean fuzzy sets are proposed.•Sensitivity analysis is performed by exchanging criteria weights with each other.•Comparative analysis is employed to validate the robustness of the method.
Humans learn to drive through both practice and theory, for example, by studying the rules, while most self‐driving systems are limited to the former. Being able to incorporate human knowledge of ...typical causal driving behavior should benefit autonomous systems. We propose a new approach that learns vehicle control with the help of human advice. Specifically, our system learns to summarize its visual observations in natural language, predict an appropriate action response (eg, “I see a pedestrian crossing, so I stop”), and predict the controls, accordingly. Moreover, to enhance the interpretability of our system, we introduce a fine‐grained attention mechanism that relies on semantic segmentation and object‐centric RoI pooling. We show that our approach of training the autonomous system with human advice, grounded in a rich semantic representation, matches or outperforms prior work in terms of control prediction and explanation generation. Our approach also results in more interpretable visual explanations by visualizing object‐centric attention maps. We evaluate our approach on a novel driving dataset with ground‐truth human explanations, the Berkeley DeepDrive eXplanation (BDD‐X) dataset.
Towards learning more human‐like driving behavior, we propose to use human advice in the form of observation‐action rules.
•Familiarity, age, education, and income were predictors of WTP for SDV technology.•Trust and perceived benefit were positive predictors of WTP.•Perceived risk and dread were negative predictors of ...WTP.•About a quarter of Chinese participants were unwilling to pay extra for SDV technology.
Research on willingness to pay (WTP) can provide practical insights for assessing the value of self-driving vehicle (SDV) technology in the vehicle market. Are people willing to pay extra for the technology? What demographic and psychological factors can influence people’s WTP for this technology? These questions are not yet well investigated. We conducted surveys in two cities in China (total N = 1355) and examined WTP and its potential demographic determinants (familiarity, age, gender, education, and income) and psychological determinants (perceived benefit and risk of SDVs, anticipated perceived dread riding in SDVs, and trust in SDVs). About 26.3% of participants were unwilling to pay extra, 39.3% were willing to pay less than $2900, and the remaining 34.3% were willing to pay more than $2900. Younger and highly educated participants with higher-income were willing to pay more. Participants who had heard about SDVs before the survey reported higher WTP and higher trust and perceived higher benefits, lower risks, and lower dread. Trust and perceived benefit were positive predictors of WTP and perceived risk and perceived dread were negative predictors of WTP. Our results may offer practical implications for increasing the public’s acceptance and WTP of SDVs.
The valet parking provides convenience for users and thus is becoming increasingly popular nowadays. However, this valet parking service incurs the risks of location information leakage and vehicle ...theft. To address these issues, in this paper, we propose a blockchain‐based privacy‐preserving valet parking protocol (B‐park) for self‐driving vehicles. The protocol guarantees the privacy of users while providing transparency and auditability of the valet parking service. We define the system model of privacy‐preserving valet parking for self‐driving vehicles in blockchain setting and formalize its security properties including anonymity, conversion blindness, nonframeability, and traceability. We provide a concrete construction of B‐park as well based on a new variant of Pointcheval‐Sanders group signature. Then, we demonstrate that our protocol is secure in the random oracle model. Finally, we evaluate the performance of B‐park to demonstrate its effectiveness and practicability.
We propose a blockchain‐based privacy‐preserving valet parking protocol called B‐park for self‐driving vehicles. The protocol guarantees the privacy of users while providing transparency and auditability of the service of valet parking.
Trajectory planning is essential for self-driving vehicles and has stringent requirements for accuracy and efficiency. The existing trajectory planning methods have limitations in the feasibility of ...planned trajectories and computational efficiency. This paper proposes a life-long learning framework to achieve effective and high-accuracy direct trajectory planning (DTP) tasks. Based on generative adversarial networks (GANs), this study develops a lightweight GDTP model to map the initial/final states and the control action sequence. Additionally, by embedding the GDTP into the rapidly-exploring random tree (RRT), a GDTP-RRT algorithm is further designed for long-distance and multi-stage planning tasks. Taking the tractor-trailer as an application case, we test the proposed method in multiple scenarios with varying characteristics. The experimental results show that the method can plan highly feasible trajectories in a short time, compared with the most applied algorithm - the cubic curve RRT* (CCRRT*). It is found that the tracking errors of our method are 29.1% and 44.1% lower than the CCRRT* in terms of position and heading angle. This paper provides an effective and stable vehicle trajectory planning method for complex self-driving tasks.
Autonomous vehicles are expected to offer a higher comfort of traveling at lower prices and at the same time to increase road capacity - a pattern recalling the rise of the private car and later of ...motorway construction. Using the Swiss national transport model, this research simulates the impact of autonomous vehicles on accessibility of the Swiss municipalities. The results show that autonomous vehicles could cause another quantum leap in accessibility. Moreover, the spatial distribution of the accessibility impacts implies that autonomous vehicles favor urban sprawl and may render public transport superfluous except for dense urban areas.
In March 2018, an Uber-pedestrian crash and a Tesla's Model X crash attracted a lot of media attention because the vehicles were operating under self-driving and autopilot mode respectively at the ...time of the crash. This study aims to conduct before-and-after sentiment analysis to examine how these two fatal crashes have affected people's perceptions of self-driving and autonomous vehicle technology using Twitter data. Five different and relevant keywords were used to extract tweets. Over 1.7 million tweets were found within 15 days before and after the incidents with the specific keywords, which were eventually analyzed in this study. The results indicate that after the two incidents, the negative tweets on “self-driving/autonomous” technology increased by 32 percentage points (from 14% to 46%). The compound scores of “pedestrian crash”, “Uber”, and “Tesla” keywords saw a 6% decrease while “self-driving/autonomous” recorded the highest change with an 11% decrease. Before the Uber-incident, 19% of the tweets on Uber were negative and 27% were positive. With the Uber-pedestrian crash, these percentages have changed to 30% negative and 23% positive. Overall, the negativity in the tweets and the percentage of negative tweets on self-driving/autonomous technology have increased after their involvement in fatal crashes. Providing opportunities to interact with this developing technology has shown to positively influence peoples' perception.
•Public perceptions play a crucial role in wide adoption of Autonomous Vehicles (AVs).•Sentiment analysis on 1.7 million tweets was conducted to study public's perceptions.•Tweets on AVs are more negative after their involvement in fatal crashes.•Safety facts on AVs should be publicized to curtail the negativity on this technology.
•Analysis of pedestrians’ informational needs towards self-driving cars.•Information on automated driving status explains absence of a driver.•Additional information on vehicle’s perception is ...perceived as obvious gimmick.•Additional information on vehicle’s intent leads to subsequent improvements.•Pedestrians show highly individual crossing and clearing strategies.
The objective of this study was to investigate pedestrians’ informational needs towards self-driving vehicles (SDVs). Previous research has shown that external human-machine interfaces (eHMIs) compensate for pedestrian-driver communication when SDVs are integrated into traffic. However, detailed insights on which information the eHMI shall provide lack so far. In a mixed design study, N = 59 participants encountered a simulated driverless vehicle in different traffic scenarios (a. unsignalized intersection vs. b. parking lot; between-subject factor). We investigated the effect of no eHMI (baseline) vs. eHMIs displaying the automated driving system (ADS) status, and informing subsequently about its perception of the pedestrian and/or its intent for the next maneuver ((1) no eHMI, (2) status eHMI, (3) status + perception eHMI, (4) status + intent eHMI, (5) status + perception + intent eHMI; within-subject factor). A mixed-methods design was used to explore participants’ subjective feelings, traffic behavior, and underlying attitudes. The findings reveal that any eHMI contributes to a more positive feeling towards SDVs compared to the baseline condition without eHMI, consistent among traffic scenarios: participants felt significantly safer, reflected greater trust and user experience ratings, and perceived the SDV as more intelligent and transparent. The status indicator mainly drives these beneficial effects on subjective measures. Participants reported that the status information explains the absence of a driver steering the vehicle. Compared to the status eHMI, the status + perception eHMI reflects no further benefit regarding subjective feelings and even has a negative impact on traffic flow. Moreover, participants regarded the additional information on the vehicle’s perception as an obvious gimmick. On the contrary, the status + intent eHMI increases user experience, perceived intelligence, and transparency for pedestrians more than the mere status eHMI. Participants reported that additionally informing about the vehicle’s intent adds a further sense of safety. The present study failed to show any improvements in traffic flow but found evidence for individual crossing and clearing strategies among pedestrians. This work can inform the future design of eHMIs.