Sparrho Insights: 3-minute summaries of cutting-edge science based on peer-reviewed research
Self-driving cars: When will it really happen? Further reading for 16 Mar 2017 issue
In 10 seconds? Detecting and protecting pedestrians is still a big challenge for fully autonomous cars and assisted driving systems. How would you feel about riding a vehicle that might sacrifice you to minimise fatalities in an emergency situation?
Not convinced? Progress in machine learning method has increased the accuracy of pedestrian detection by leaps and bounds (see below + read more), but the question of ethics behind the driving systems remains unresolved.
What's the dilemma?
Who to kill – Imagine the dilemma during an emergency: should a driverless car swerve out of the way of pedestrians at the expense of driving into a concrete wall and killing its passengers? Manufacturers of self-driving cars are in essence predetermining the survival of their riders through programming decisions they make now. (read more)
Nobody wants to say – Mercedes Benz is the only company so far to have made a public statement, but quickly reversed their position after a manager's interview was sensationalised into headlines like Mercedes's Self-Driving Cars Will Kill Pedestrians Over Drivers.
Given all this talk of ethics, when will fully autonomous cars hit the road?
Singapore is pioneering – In August 2016, MIT software spinout nuTonomy partnered with ride-hail platform Grab to launch the world’s first autonomous taxis in Singapore, with the aim to create a full fleet by 2018. (read more)
So a taxi with no human at all? – Well not yet, in this current test phase, a human engineer still sits in the front seat on every ride, prepared to take the wheel if necessary, while a backseat researcher monitors the vehicle's computers. (read more)
Abstract: This article describes an automated sensor-based system to detect pedestrians in an autonomous vehicle application. Although the vehicle is equipped with a broad set of sensors, the article focuses on the processing of the information generated by a Velodyne HDL-64E LIDAR sensor. The cloud of points generated by the sensor (more than 1 million points per revolution) is processed to detect pedestrians, by selecting cubic shapes and applying machine vision and machine learning algorithms to the XY, XZ, and YZ projections of the points contained in the cube. The work relates an exhaustive analysis of the performance of three different machine learning algorithms: k-Nearest Neighbours (kNN), Naïve Bayes classifier (NBC), and Support Vector Machine (SVM). These algorithms have been trained with 1931 samples. The final performance of the method, measured a real traffic scenery, which contained 16 pedestrians and 469 samples of non-pedestrians, shows sensitivity (81.2%), accuracy (96.2%) and specificity (96.8%).
Pub.: 28 Dec '16, Pinned: 07 Mar '17
Abstract: Abstract Recently, pedestrian detection systems have become an important technology in the development of the advanced driver assistance system (ADAS) for the autonomous car. The histogram of oriented gradients (HOG) is currently the most basic algorithm for detecting pedestrians, but it treats the entire body of the pedestrian as one single feature. In other words, if the entire body of the pedestrian is not visible, the detection rate under HOG decreases markedly. To solve this problem, we propose a detection system using a deformable part model (DPM) that divides the pedestrian data into two parts using a latent support vector machine (SVM)-based machine-learning technique. Experimental results show that our approach achieves better performance in a detection system than the existing method. In practice, there are many occlusions in the environment in front of the vehicle. For example, the surrounding transport facilities, such as a car or another obstacle, can occlude a pedestrian. These occlusions can increase the false detection rate and cause difficulties during the detection process. Our proposed method uses a different approach and can easily be applied in real-world scenarios, regardless of occlusions.AbstractRecently, pedestrian detection systems have become an important technology in the development of the advanced driver assistance system (ADAS) for the autonomous car. The histogram of oriented gradients (HOG) is currently the most basic algorithm for detecting pedestrians, but it treats the entire body of the pedestrian as one single feature. In other words, if the entire body of the pedestrian is not visible, the detection rate under HOG decreases markedly. To solve this problem, we propose a detection system using a deformable part model (DPM) that divides the pedestrian data into two parts using a latent support vector machine (SVM)-based machine-learning technique. Experimental results show that our approach achieves better performance in a detection system than the existing method. In practice, there are many occlusions in the environment in front of the vehicle. For example, the surrounding transport facilities, such as a car or another obstacle, can occlude a pedestrian. These occlusions can increase the false detection rate and cause difficulties during the detection process. Our proposed method uses a different approach and can easily be applied in real-world scenarios, regardless of occlusions.
Pub.: 25 Oct '16, Pinned: 11 Mar '17
Abstract: The detection of small road hazards, such as lost cargo, is a vital capability for self-driving cars. We tackle this challenging and rarely addressed problem with a vision system that leverages appearance, contextual as well as geometric cues. To utilize the appearance and contextual cues, we propose a new deep learning-based obstacle detection framework. Here a variant of a fully convolutional network is used to predict a pixel-wise semantic labeling of (i) free-space, (ii) on-road unexpected obstacles, and (iii) background. The geometric cues are exploited using a state-of-the-art detection approach that predicts obstacles from stereo input images via model-based statistical hypothesis tests. We present a principled Bayesian framework to fuse the semantic and stereo-based detection results. The mid-level Stixel representation is used to describe obstacles in a flexible, compact and robust manner. We evaluate our new obstacle detection system on the Lost and Found dataset, which includes very challenging scenes with obstacles of only 5 cm height. Overall, we report a major improvement over the state-of-the-art, with relative performance gains of up to 50%. In particular, we achieve a detection rate of over 90% for distances of up to 50 m. Our system operates at 22 Hz on our self-driving platform.
Pub.: 20 Dec '16, Pinned: 11 Mar '17
Abstract: Suppose that a driverless car is headed toward five pedestrians. It can stay on course and kill them or swerve into a concrete wall, killing its passenger. On page 1573 of this issue, Bonnefon et al. (1) explore this social dilemma in a series of clever survey experiments. They show that people generally approve of cars programmed to minimize the total amount of harm, even at the expense of their passengers, but are not enthusiastic about riding in such “utilitarian” cars—that is, autonomous vehicles that are, in certain emergency situations, programmed to sacrifice their passengers for the greater good. Such dilemmas may arise infrequently, but once millions of autonomous vehicles are on the road, the improbable becomes probable, perhaps even inevitable. And even if such cases never arise, autonomous vehicles must be programmed to handle them. How should they be programmed? And who should decide? Author: Joshua D. Greene
Pub.: 24 Jun '16, Pinned: 07 Mar '17
Abstract: Publication date: Available online 15 September 2016 Source:Regional Science and Urban Economics Author(s): Roman Zakharenko The effects of autonomous vehicles (AVs) on urban forms are modeled, calibrated, and analyzed. Vehicles are used for commute between peripheral home and central work, and require land for parking. An advantage of AVs is that they can optimize the location of day parking, relieving downtown land for other uses. They also reduce the per-kilometer cost of commute. Increased AV availability increases worker welfare, traffic, travel distances, and the city size. Land rents increase in the center but decrease in the periphery. Possible locations of AV daytime parking are analyzed. The effects of AV introduction on traffic and on mass transit coverage are discussed.
Pub.: 17 Sep '16, Pinned: 07 Mar '17
Abstract: This study investigates the challenges and opportunities pertaining to transportation policies that may arise as a result of emerging autonomous vehicle (AV) technologies. AV technologies can decrease the transportation cost and increase accessibility to low-income households and persons with mobility issues. This emerging technology also has far-reaching applications and implications beyond all current expectations. This paper provides a comprehensive review of the relevant literature and explores a broad spectrum of issues from safety to machine ethics. An indispensable part of a prospective AV development is communication over cars and infrastructure (connected vehicles). A major knowledge gap exists in AV technology with respect to routing behaviors. Connected-vehicle technology provides a great opportunity to implement an efficient and intelligent routing system. To this end, we propose a conceptual navigation model based on a fleet of AVs that are centrally dispatched over a network seeking system optimization. This study contributes to the literature on two fronts: (i) it attempts to shed light on future opportunities as well as possible hurdles associated with AV technology; and (ii) it conceptualizes a navigation model for the AV which leads to highly efficient traffic circulations.
Pub.: 29 Aug '16, Pinned: 07 Mar '17
Abstract: With the potential to save nearly 30 000 lives per year in the United States, autonomous vehicles portend the most significant advance in auto safety history by shifting the focus from minimization of postcrash injury to collision prevention. I have delineated the important public health implications of autonomous vehicles and provided a brief analysis of a critically important ethical issue inherent in autonomous vehicle design. The broad expertise, ethical principles, and values of public health should be brought to bear on a wide range of issues pertaining to autonomous vehicles. (Am J Public Health. Published online ahead of print February 16, 2017: e1-e6. doi:10.2105/AJPH.2016.303628).
Pub.: 17 Feb '17, Pinned: 07 Mar '17
Abstract: The USA has the worst motor vehicle safety problem among high-income countries and is pressing forward with the development of autonomous automobiles to address it. Government guidance and regulation, still inadequate, will be critical to the safety of the public. The analysis of this public health problem in the USA reveals the key factors that will determine the benefits and risks of autonomous vehicles around the world.
Pub.: 25 Jan '17, Pinned: 07 Mar '17
Abstract: A recent trend in the luxury industry shows many companies using brand extension strategies to leverage their assets among competitors. Despite the popularity of brand extensions, limited research has been conducted to determine its effectiveness when using parent product brands in order to introduce new service brands. Thus, the primary purpose of this study is to propose the framework to increase the understanding of luxury brand extensions by focusing on the horizontal category brand extension from product to service. The impact of perceived quality, brand association and brand loyalty of the luxury car brand on consumers’ perception and attitude toward the rental car service brand was observed to differentiate from other research that focuses on brand extension within product categories. A total of 324 samples were collected and analyzed using structural equation modeling with AMOS. The brand association of the parent brand showed a significant impact on the evaluation of the extended service brand. In addition, a high level of the brand association was also found to influence the final purchase decision of the extended service brand. However, this study could not find any significant effect of perceived quality and brand loyalty for the parent product brand on the extended service brand. The framework proposed in this study has merit to increase the understanding of service brand extensions by exploring luxury car brands and luxury rental car brands. In the service brand extension, there was a gap in consumer perception between the product brand and the extended service brand. In order to maximize the positive impact of the parent brand, marketers and retailers should investigate the different roles of the brand equity items.
Pub.: 25 Jul '16, Pinned: 16 Mar '17
Abstract: Abstract The possibility of using data from multiplex networks on vehicles in road tests, in the development of intelligent transportation systems, and in control systems for autonomous (self-driving) vehicles is considered.AbstractThe possibility of using data from multiplex networks on vehicles in road tests, in the development of intelligent transportation systems, and in control systems for autonomous (self-driving) vehicles is considered.
Pub.: 01 Oct '16, Pinned: 07 Mar '17
Abstract: To lay the basis of studying autonomous driving comfort using driving simulators, we assessed the behavioral validity of two moving-base simulator configurations by contrasting them with a test-track setting.With increasing level of automation, driving comfort becomes increasingly important. Simulators provide a safe environment to study perceived comfort in autonomous driving. To date, however, no studies were conducted in relation to comfort in autonomous driving to determine the extent to which results from simulator studies can be transferred to on-road driving conditions.Participants ( N = 72) experienced six differently parameterized lane-change and deceleration maneuvers and subsequently rated the comfort of each scenario. One group of participants experienced the maneuvers on a test-track setting, whereas two other groups experienced them in one of two moving-base simulator configurations.We could demonstrate relative and absolute validity for one of the two simulator configurations. Subsequent analyses revealed that the validity of the simulator highly depends on the parameterization of the motion system.Moving-base simulation can be a useful research tool to study driving comfort in autonomous vehicles. However, our results point at a preference for subunity scaling factors for both lateral and longitudinal motion cues, which might be explained by an underestimation of speed in virtual environments.In line with previous studies, we recommend lateral- and longitudinal-motion scaling factors of approximately 50% to 60% in order to obtain valid results for both active and passive driving tasks.
Pub.: 23 Dec '16, Pinned: 07 Mar '17
Abstract: Through a review of long-range transportation plans and interviews with planners, this article examines how large metropolitan planning organizations are preparing for autonomous vehicles. In just a few years, the prospect of commercially available self-driving cars and trucks has gone from a futurist fantasy to a likely near-term reality. However, uncertainties about the new technology and its relationship to daily investment decisions have kept mention of self-driving cars out of nearly all long-range transportation plans. Nevertheless, interviewees are keeping a close watch on the new technology and actively looking to understand and plan for future impacts.
Pub.: 02 Jun '16, Pinned: 07 Mar '17
Abstract: Detecting small obstacles on the road ahead is a critical part of the driving task which has to be mastered by fully autonomous cars. In this paper, we present a method based on stereo vision to reliably detect such obstacles from a moving vehicle. The proposed algorithm performs statistical hypothesis tests in disparity space directly on stereo image data, assessing freespace and obstacle hypotheses on independent local patches. This detection approach does not depend on a global road model and handles both static and moving obstacles. For evaluation, we employ a novel lost-cargo image sequence dataset comprising more than two thousand frames with pixelwise annotations of obstacle and free-space and provide a thorough comparison to several stereo-based baseline methods. The dataset will be made available to the community to foster further research on this important topic. The proposed approach outperforms all considered baselines in our evaluations on both pixel and object level and runs at frame rates of up to 20 Hz on 2 mega-pixel stereo imagery. Small obstacles down to the height of 5 cm can successfully be detected at 20 m distance at low false positive rates.
Pub.: 15 Sep '16, Pinned: 07 Mar '17
Abstract: This paper discusses opportunities to parallelize graph based path planning algorithms in a time varying environment. Parallel architectures have become commonplace, requiring algorithm to be parallelized for efficient execution. An additional focal point of this paper is the inclusion of inaccuracies in path planning as a result of forecast error variance, accuracy of calculation in the cost functions and a different observed vehicle speed in the real mission than planned. In this context, robust path planning algorithms will be described. These algorithms are equally applicable to land based, aerial, or underwater mobile autonomous systems. The results presented here provide the basis for a future Research project in which the parallelized algorithms will be evaluated on multi and many core systems such as the dual core ARM Panda board and the 48 core Single-chip Cloud Computer (SCC). Modern multi and many core processors support a wide range of performance vs. energy tradeoffs that can be exploited in energyconstrained environments such as battery operated autonomous underwater vehicles. For this evaluation, the boards will be deployed within the Slocum glider, a commercially available, buoyancy driven autonomous underwater vehicle (AUV).
Pub.: 26 Feb '17, Pinned: 07 Mar '17
Abstract: Autonomous vehicles are becoming an essential tool in a wide range of environmental applications that include ambient data acquisition, remote sensing, and mapping of the spatial extent of pollutant spills. Among these applications, pollution source localization has drawn increasing interest due to its scientific and commercial interest and the emergence of a new breed of robotic vehicles capable of operating in harsh environments without human supervision. The aim is to find the location of a region that is the source of a given substance of interest (e.g. a chemical pollutant at sea or a gas leakage in air) using a group of cooperative autonomous vehicles. Motivated by fast paced advances in this challenging area, this paper surveys recent advances in searching techniques that are at the core of environmental monitoring strategies using autonomous vehicles.
Pub.: 04 Mar '17, Pinned: 07 Mar '17