Autonomous Taxi Research Articles

Process of Learning from Demonstration with Paraconsistent Artificial Neural Cells for Application in Linear Cartesian Robots

Paraconsistent Annotated Logic (PAL) is a type of non-classical logic based on concepts that allow, under certain conditions, for one to accept contradictions without invalidating conclusions. The Paraconsistent Artificial Neural Cell of Learning (lPANCell) algorithm was created from PAL-based equations. With its procedures for learning discrete patterns being represented by values contained in the closed interval between 0 and 1, the lPANCell algorithm presents responses similar to those of nonlinear dynamical systems. In this work, several tests were carried out to validate the operation of the lPANCell algorithm in a learning from demonstration (LfD) framework applied to a linear Cartesian robot (gantry robot), which was moving rectangular metallic workpieces. For the LfD process used in the teaching of trajectories in the x and y axes of the linear Cartesian robot, a Paraconsistent Artificial Neural Network (lPANnet) was built, which was composed of eight lPANCells. The results showed that lPANnet has dynamic properties with a robustness to disturbances, both in the learning process by demonstration, as well as in the imitation process. Based on this work, paraconsistent artificial neural networks of a greater complexity, which are composed of lPANCells, can be formed. This study will provide a strong contribution to research regarding learning from demonstration frameworks being applied in robotics.

Living in a City Where Automated Taxis are Operating and Using Them: Does This Affect Consumers’ Preferences?

In this paper we study the impact of having used an automated taxi (AT) or simply living in a city where ATs are operating on users’ preferences for and attitudes toward ATs. This paper aims then to contribute to the research on ATs and on the impact of experiencing innovations. Data were collected in three major cities in China where ATs were in operation (AT cities), and other major cities where only normal taxis (NTs) were operating (NT cities). A stated choice (SC) experiment was built that included level-of-services attributes, specific AT in-vehicle features, levels of AT adoption and customers’ reviews. In addition, a set of attitudinal statements was included to measure injunctive norms, hedonic motivation and trust. Hybrid choice models were estimated accounting for intra-individual correlation. Results show that living in a city where AT are operating does not affect the willingness to pay for travel and waiting time, which is an expected result. But those living in AT cities compared with those living in NT cities are willing to pay just half the amount to have the in-vehicle feature “change destination during the trip” and around three times as much for good reviews about ATs for trips less than 30 min. Interestingly, trust has a positive and significant impact on the choice of AT only for those living in NT cities, while hedonic motivation is significant only for those living in AT cities. Finally, we did not find differences between those who had used an AT and those who had never used it.

Understanding User Acceptability Towards to Robo Taxi Based on Value Based Adoption Model

Understanding User Acceptability Towards to Robo Taxi Based on Value Based Adoption Model

Automatic offline path planning of robots grinding multi-curved surfaces on large ship propellers – A human-in-the-loop approach

Multi-curved components represent a particular challenge when machining with robots, as the machining is subject to a wide range of technical limits depending on the robot system used. An example of such a multi-curved component can be found in the ship propeller, which is one of the most important components of a ship and represents a relatively complex part in the production process. Manufacturing in the casting process is resource- and time-consuming and requires extensive finishing, which also includes grinding to final size. Technologically, this operation is still performed manually or mechanically assisted and requires employees with a high level of experience in the operation and handling of grinding technology. In order to prevent the increasing shortage of skilled workers and to make the finishing process independent from employee-specific knowledge, it is necessary to develop an automatic grinding process. In this context, offline path planning for multi-curved surfaces has emerged as a particular challenge, since existing approaches do not sufficiently consider the robot axis positions nor the pose limits of the end effectors. This work starts by presenting the current state of research and then introduces a human-in-the-loop approach that solves the problem of 3D trajectory planning using two-dimensional viewing planes. The trajectories thus found are projected onto the surface of the propeller and analyzed for potential collisions with the robot, grinding efficiency and reachability. Furthermore, it is possible to calculate an automatic removal characteristic by overlaying a simplified casting model and ideal model of a propeller.

Moving toward a More Sustainable Autonomous Mobility, Case of Heterogeneity in Preferences

Autonomous vehicles (AVs) have a number of potential advantages, although some research indicates that this technology may increase dependence on private cars. An alternative approach to bringing such technology to market is through autonomous taxis (ATs) and buses, which can assist in making transportation more sustainable. This paper aims at examining the role of attitudinal, travel-related, and individual factors in preferences for a modal shift from conventional cars toward ATs and exclusive-lane autonomous buses (ELABs), exploring the existence of heterogeneity and its possible sources. The proposed mixed logit model with a decomposition of random coefficients uses 1251 valid responses from a stated preference survey distributed in Tehran, in 2019. Results show that there is significant taste variation among individuals with respect to ATs’ travel costs, ELABs’ travel times, and walking distances to ELAB stations. Furthermore, exploring the sources of heterogeneity indicates that women are more sensitive to ATs’ travel costs and walking distances to ELAB stations while they are less sensitive to ELABs’ travel times. Moreover, travel time in discretionary activities reduces the utility of ELABs more than it does in mandatory activities. Transportation authorities can use these findings to establish more effective policies for the successful implementation of AVs.

Equilibrium analysis of trip demand for autonomous taxi services in Nagoya, Japan

Equilibrium analysis of trip demand for autonomous taxi services in Nagoya, Japan

RiderAuth: A cancelable touch-signature based rider authentication scheme for driverless taxis

Driverless taxis are now closer to their worldwide launch, however, unsupervised physical access to the riders can pose unexpected safety and security risks to the connected and autonomous vehicle’s ecosystems. Thus, the need for human-to-machine authentication in driverless settings cannot be disregarded to maintain their smooth operations. In this paper, we propose a first rider authentication scheme, RiderAuth, using touch-signature behavioral biometrics to authenticate customers before they commence their ride in driverless settings. We also propose a template protection scheme for finite length touch-signature features using a random orthogonal projection method to address biometric template leakage. We design Android-based client and Cloud-based server prototypes to evaluate both the security and usability of our proposed scheme. We collect a new database of 1800 touch-signatures samples from 30 users in an unsupervised field experiment for two weeks. RiderAuth achieves a True Acceptance Rate (TAR) of 96.66% at False Acceptance Rate (FAR) of 0.01% without template protection and a TAR of 94.88% at FAR=0.01% with proposed template protection scheme. Subsequently, we implement a d-prime weighted score-level fusion to further enhance RiderAuth’s accuracy. We evaluate the unlinkability and revocability of the proposed template protection scheme to verify its suitability for authenticating riders in a privacy preserved setting. In addition, we conduct a survey to evaluate user acceptance of our proposed scheme using the System Usability Scale (SUS) where we get a score of ≈70% suggesting a positive experience and acceptance of RiderAuth.

The effect of increasing vehicle utilization on the automotive industry

Shared mobility is widely expected to play an important role in the future of transportation. Sharing vehicles (using services such as ride-hailing, peer-to-peer car-sharing, and autonomous taxis) will allow people to enjoy the benefits of automobile use without ownership, access various types of mobility services on-demand, and create value by increasing the utilization of these expensive and durable assets. Most analysts agree that widespread adoption of shared mobility would cause the size of the on-road automobile fleet to shrink, potentially dramatically, because the same amount of personal mobility can be provided by fewer vehicles. There is less agreement, however, on the effect higher utilization will have on the rate of new vehicle sales: some believe that vehicle sales will fall similarly, while others believe there will be no change in sales, or even an increase in sales as fleets of shared vehicles turn over more frequently. In this paper, we seek to clarify the effect that emerging mobility technologies will have on the future rate of new vehicle sales in the United States, modeling how the sales rate varies with factors such as population growth, vehicle utilization, and vehicle durability. We show across a range of plausible scenarios that vehicle sales are likely to remain steady or increase in coming decades. However, the potential exists for a temporary surge or dip in sales as the composition of new vehicle sales transitions, requiring effective mental models if managers are to make efficient production and capacity planning decisions during this time.

Willingness to pay for automated taxis: a stated choice experiment to measure the impact of in-vehicle features and customer reviews

This study aims to contribute to the limited empirical evidence on Automated Taxis (ATs) by analysing the in-vehicle features that ATs ought to have to satisfy taxi passengers requests that in a Normal Taxi (NT) are dealt with a direct communication between the passenger and the driver. Based on the results of focus groups and several pilots, two in-vehicle features are tested: change of destination and chat with an operator during the trip. The paper also contributes to the body of literature on the impact of social influence on the adoption of innovation by testing the impact of consumer reviews other than the typical measure of adoption and injunctive norms. A Stated Choice (SC) experiment was built putting particular effort in the definition and presentation of the new attributes tested. The study was applied in China among current users of NTs (i.e. with the driver). Hybrid choice models were estimated and a resampling technique was used to test the model sensitivity to the sample gathered. Results show that on average, Chinese taxi users are willing to pay 0.35 Euros to have the option to ‘change the destination’ and 0.78 Euros to be able to ‘chat with an operator’ inside the AT. Among the social influence attributes, the reviews from previous customers confirmed to be the most effective measure, users are willing to pay 1.58 Euros more to use a taxi that got good reviews for long trips (≥ 30 min) and 0.57 Euros for shorter trips. The Willingness to Pay (WTP) estimated are all significant at more than 95% and have a narrow confidence intervals.

BBARHS: Blockchain-Based Anonymous Ride-Hailing Scheme for Autonomous Taxi Network

In the past few years, ride-hailing platforms such as Uber, Waymo, and Baidu have built their own autonomous taxi system. Unlike public transit services, ride-hailing platforms raise severe privacy issues. To provide excellent autonomous taxi service, some significant security and privacy problems must be addressed. In this study, we present the security and privacy threats and first proposed blockchain-based anonymous ride-hailing scheme (BBARHS) for autonomous taxi network. We give the formal system model and security model of BBARHS. Then, we outline the concrete BBARHS scheme by making use of Monero and some efficient crypto tools. Through security analysis and performance analysis, the designed scheme is provably secure and efficient. The analysis results also show the designed BBARHS scheme is practical for autonomous taxi network.

Review of shared online hailing and autonomous taxi services

As the technology of autonomous vehicle develops, online hailing autonomous taxi system is regarded as one of the most popular public transportation services in the future. Studies related to demand forecasting, ride matching, path planning, relocation, and pricing strategy for shared online hailing and autonomous taxi services have emerged in recent years. In this study, we conducted a survey based on 141 representative literatures from 1995 to 2022 to understand the state-of-the-art of the key problems of operating autonomous taxi service. First, a comprehensive review of the components of the shared autonomous taxi modelling is presented. Then, how the emerging technologies such as internet of vehicles, big data, cloud and edge computing, and blockchain can be used to enhance the autonomous taxi service is discussed. Last, the current research challenges and the concern or hurdle in public’s adoption of autonomous taxi services are identified.

General Procedure for Servo-Axis Design in Multi-Degree-of-Freedom Machinery Subject to Mixed Loads

The motion performances of multi-axis machinery are strongly related to the proper design of the servo axes driving the actuated joints, which must fulfil motion requests in terms of speed and torque for the motion design task. Multi-degree-of-freedom machinery highlights, at the joint level, mixed-load conditions due to high dynamic movements and external working forces. Moreover, such systems have a dynamic coupling between the axes due to their mechanical configuration. This mixed-load condition makes the servo-axes design, whose result is the selection of the couple power drive system and transmission, not a trivial procedure. This paper proposes an extensive theoretical discussion that leads to a general procedure for the direct design of the servo axes of multi-degree-of-freedom machinery subject to a mixed-load condition and dynamic coupling between the mechanical joints. The proposed procedure starts with identifying the loads required on the actuated joints without distinguishing between the different contributions. Then, in a few steps, the methodology leads to the direct choice of the PDS/transmission units through a single graphical representation where the load’s peak, thermal and speed conditions are highlighted, along with the possible PDSs and transmissions. The graphical representation proposed also allows the easy and efficient analysis of the effect of transmission selection on the torque and speed margins. Finally, the paper presents the effectiveness of the proposed procedure, applied to the design of the servo axes of a linear delta robot.

Autonomous Taxi Driving Environment Using Reinforcement Learning Algorithms

Autonomous Taxi Driving Environment Using Reinforcement Learning Algorithms

An attention enhanced dilated CNN approach for cross-axis industrial robotics fault diagnosis

An industrial robot is a complex mechatronics system, whose failure is hard to diagnose based on monitoring data. Previous studies have reported various methods with deep network models to improve the accuracy of fault diagnosis, which can get an accurate prediction model when the amount of data sample is sufficient. However, the failure data is hard to obtain, which leads to the few-shot issue and the bad generalization ability of the model. Therefore, this paper proposes an attention enhanced dilated convolutional neural network (D-CNN) approach for the cross-axis industrial robotics fault diagnosis method. Firstly, key feature extraction and sliding window are adopted to pre-process the monitoring data of industrial robots before D-CNN is introduced to extract data features. And self-attention is used to enhance feature attention capability. Finally, the pre-trained model is used for transfer learning, and a small number of the dataset from another axis of the multi-axis industrial robot are used for fine-tuning experiments. The experimental results show that the proposed method can reach satisfactory fault diagnosis accuracy in both the source domain and target domain.

Assessment of ISO Standardisation to Identify an Industrial Robot's Base Frame

• BS EN ISO 9283:1998 and ISO/TR 13,309:1995 require a definitive method on how to construct the robot base frame with modern equipment and associated metrology software. • Bespoke methods in the literature can vary the robot base frame position, when repeated. • A design of experiments based method is presented that investigates if factors used in the robot base frame construction affect the repeatability of the process, and if they have a statistically significant influence on the outcome. There is a growing need for industrial robots to undertake high tolerance operations in line with Industry 4.0 (I4.0) demands and requirements. This requires robotic accuracy to be understood and quantified within appropriate bounds, relative to their application area. One such area is complex aerospace assembly where robotic accuracy is of particular importance if the full potential and benefits of I4.0 are to be realised in the sector. The standards BS EN ISO 9283:1998 and ISO/TR 13,309:1995 outline the requirements for calculating an industrial robot's accuracy. The standards highlight equipment capable of achieving this and they note that the robot base frame (RBF) must be determined as part of the process. The RBF is an exact position within the robot, and therefore it should be established with both accuracy and precision. However, a specific process or approach is not provided in either standard for the determination of a RBF. This has resulted in the use of bespoke methods by various researchers for the determination of their RBF using metrological equipment and associated software. These ad-hoc methods are not globally applied and the rationale and justification for their use remains unpublished. Previous research that presented a process used to construct the RBF, resulted in a varying RBF origin position when repeated. The work presented in this paper provides the basis for a common approach to the determination of a RBF which integrates metrology hardware with a Design of Experiments (DOE) approach to select an appropriate measurement routine. The DOE approach investigates how different factors (e.g. robot axis used, number of point positions used, their positions, and repetitions of their occurrence) influence the repeatability in establishing the RBF, by measuring the positions of points that the robot attains using different levels for each factor. This study used a Universal Robot to develop and demonstrate the proposed method for RBF determination using the factors that were found to affect point repeatability. This new method was validated by comparing the outcome of the applied process to four methods that used random combinations of factors. The approach was found to increase the repeatability in establishing the RBF origin point by 93.4%, compared to a previous method that used arbitrarily chosen factors.

Design and implementation of a highly integrated dual hemisphere capsule robot.

To achieve cancer screening in any appointed position in 3D regions of the gastrointestinal (GI) tract such as esophagus, stomach and colon, a highly integrated dual hemisphere capsule robot (DHCR) with a novel three-layer nested structure is proposed. Based on tracking effect, in which the robotic axis is likely to be approximately coincident with the orientation of thespace universal rotating magnetic field (SURMF) using the gyroscope dynamic balance, the dual hemisphere structure realizes the observation at a fixed-point in the passive mode and the rolling locomotion in the active mode by the dynamic posture control of the SURMF manipulation. The image acquisition module, wireless transmission module and driving actuator are tuned in a spherical structure, making the DHCR more compact and less invasive. To verify the maneuverability of the innovative DHCR both for observation at a fixed-point and navigation in curved intestine by aid of image, experiments are conducted in the simulated GI tract environment. The results show that the DHCR achieves effective conversion between posture adjustment and rolling locomotion, which lays a foundation for all-over inspection and medical operation inside 3D regions of the GI tract of human body.

The formation of initial trust by potential passengers of self-driving taxis

ABSTRACT Despite a current level of uncertainty, taxi service companies are incorporating autonomous cars in their services. This study examines factors influencing consumers' initial trust in autonomous taxis to extend trust model research and develop a new model in autonomous taxis. The new model examines the influence of trust base factors and subjective norms on three constructs. It assesses eight trust base factors of cognitive, personality, calculative, and institutional categories. Findings indicate that faith in intelligent machines, calculative cost/benefit perception, perceived societal benefits, organizational situation normality, and technology structural assurance are potential indicators of consumers' trusting beliefs in this context. Data was collected from undergraduate students and future research should test the model with other populations to determine if differences among groups exist. The study makes significant contributions to both theory and practice by extending the initial trust model and examining trusting factors in the context of autonomous taxis.

Explaining expected non-shared and shared use of driverless cars in Edinburgh

The rollout of driverless cars (DC) may reshape human mobility drastically. Urban roads may face a surge of car traffic if DC technologies continue to enhance the use of private cars. Shared DC use can remedy this by reducing the number of urban trips that non-shared-use DC could otherwise make. While recent research has proven the importance of socioeconomic factors on shared DC use, the influence of present sharing behaviour, personality traits, and social norms on shared DC use has not been extensively explored to date. To address this gap in this study, we employed a panel of 500 Edinburgh-based respondents through an online survey to examine the likelihood of accepting the non-shared-use and shared-use DC options for regular urban trips. We have collected data on respondents’ present carsharing and ridesharing behaviour, personality traits, social norms, and sociodemographic characteristics. To elicit the impact of these factors on the likelihood of accepting non-shared-use and shared-used DC options, ordered probit models were estimated. The model findings imply that frequent household car users and those influenced by social expectations to preserve the environment are willing to use non-shared-use DC. In contrast, city centre dwellers, cooperative and younger adults with sharing attitudes show a higher tendency towards shared DC. High-earning, working-aged and young respondents are more inclined to use a driverless taxi, whereas city-centre dwellers and those influenced by social expectations to share personal resources are more favourable towards ridesharing with a stranger in DC. These results can assist the policymakers and transport planners shape policies for promoting shared DC use and transport service providers to deliver and operate shared DC fleets efficiently.

Economic evaluation of autonomous passenger transportation services: a systematic review and meta-analysis of simulation studies

In order to prepare for their deployment, mobility simulations have been used since the mid-2010s to investigate the effects of Autonomous Vehicles (AV) services. The aim of this study is to provide a review of which impacts were considered in simulation studies, to what extent, and for which results, with an emphasis on economic impacts. Taking cost-benefit analysis (CBA) as a reference framework for the economic evaluation of mobility services, we first establish a list of standard impact indicators in CBA studies. We carry out a descriptive statistical analysis to investigate the use of these indicators, completed by a meta-analysis for the indicators that received sufficient coverage from the literature. The results show evidence of strong use of performance and operational indicators in the simulation literature, but also expose a lack of in-depth analysis concerning environmental and social indicators. With regard to the intensity of impacts, the introduction of autonomous taxis should increase vehicle- kilometers traveled (VKT) by +23% and travel time by +17% compared to conventional private cars. At the same time, it could help to cut 17% of the present fleet size. This performance could be improved if autonomous taxis were shared, limiting the increase of VKT to +6% at the cost of only slightly longer travel times (+20% instead of +17%), while also increasing fleet reduction by up to -55%. Compared to conventional bus or train lines, these on-demand AV services could permit travel time to be reduced by half and could also reduce financial operating costs. On the other hand, they would increase fleet size and VKT by three to seven times their initial values. AV impacts could vary considerably depending on the socioeconomic landscape of the implementation area. Our findings call for evaluating AV services from a wider perspective than operational and financial prisms alone as is currently the case, such as using cost-benefit analysis.JEL classification: H43, L92, L91.

Exploring the Influence of Driver's Presence on Passengers’ Willingness Level to Take Autonomous Taxis Using Enhanced Multiclass Random Forest Algorithm

Exploring the Influence of Driver's Presence on Passengers’ Willingness Level to Take Autonomous Taxis Using Enhanced Multiclass Random Forest Algorithm