Life Cycle Assessment of Connected and Automated Vehicles: Sensing and Computing Subsystem and Vehicle Level Effects

Reduction of life-cycle CO2 emissions by expanding car-sharing services: A case study on Japan

Carbon neutrality is a growing concern for all global economies. We considered the number of new and used cars registered during 2009-2018 in Japan and estimated the total number of private and shared cars, assuming that when owners abandoned their old cars, a certain percentage of the owners chose to use a car-sharing service (i.e., car rental service), instead of buying a new private car. We estimated the CO2 emissions generated during the manufacturing, driving, and disposal stages of cars, to analyze the impact of car sharing on CO2 emissions. Then, we determined the changes in the life-cycle CO2 emissions of all the cars for three car-sharing penetration rates (0, 5, and 100%), assuming that all the cars were gasoline-powered. Additionally, we analyzed how electric vehicles can optimize the proposed strategy. An increase in car-sharing services significantly reduced vehicular CO2 emissions; the decrease in CO2 emissions from private cars when owners switched to car services significantly exceeded the increase in the CO2 emissions associated with the increased number of cars. The proposed model can serve as a reliable framework to analyze the current status of CO2 emissions and simulate the future changes in car-sharing services.

Autonomous driving controllers with neuromorphic spiking neural networks

Autonomous driving is one of the hallmarks of artificial intelligence. Neuromorphic (brain-inspired) control is posed to significantly contribute to autonomous behavior by leveraging spiking neural networks-based energy-efficient computational frameworks. In this work, we have explored neuromorphic implementations of four prominent controllers for autonomous driving: pure-pursuit, Stanley, PID, and MPC, using a physics-aware simulation framework. We extensively evaluated these models with various intrinsic parameters and compared their performance with conventional CPU-based implementations. While being neural approximations, we show that neuromorphic models can perform competitively with their conventional counterparts. We provide guidelines for building neuromorphic architectures for control and describe the importance of their underlying tuning parameters and neuronal resources. Our results show that most models would converge to their optimal performances with merely 100-1,000 neurons. They also highlight the importance of hybrid conventional and neuromorphic designs, as was suggested here with the MPC controller. This study also highlights the limitations of neuromorphic implementations, particularly at higher (> 15 m/s) speeds where they tend to degrade faster than in conventional designs.

Life cycle inventory data of agricultural tractors

Life Cycle Assessments (LCA) of agricultural systems are performed using inventory data from several databases. The inventory data used for agricultural machinery and especially agricultural tractors in these databases are based on old data (from 2002 and not updated since) using trucks ("lorry") as a proxy for the manufacture of tractors. In consequences, they do not reflect the current technology used by farmers and do not allow comparison with new technologies in used in farms such as agricultural robots. The dataset proposed in this paper presents two updated Life Cycle Inventory (LCI) of an agricultural tractor. Data were collected based on the technical system of a tractor manufacturer, scientific and technical literature as well as expert opinion. Data on weight, composition, lifetime and maintenance hours of each tractor component as well as electronic parts, converter catalyst and lead battery are produced. The inventory is calculated based on the raw materials needed for the tractor manufacturing and maintenance over its lifetime as well as the energy and infrastructure needed for manufacturing. Calculations were made based on a tractor of 7300 kg with the following characteristics: 155 CV, 6 cylinders, four-wheel drive. The tractor modelled is representative of tractors from the same power category (i.e. between 100 and 199 CV and 70% of the annual sales in France). Two LCI are produced: a LCI for a 7200 h lifetime tractor, representative of an accounting depreciation, and a LCI for a 12000 h lifetime tractor, representative of the whole service life of the tractor (first use to final disposal). The functional unit is 1 kg of tractor (kg) or 1 piece (p) of tractor during its lifetime.

Barriers to the sustainable adoption of autonomous vehicles in developing countries: A multi-criteria decision-making approach

The acceptance of AI-based intelligent transportation systems requires addressing the existing barriers and the adoption of macro-decisions and policies by policymakers and governments. This study evaluates the potential barriers to the adoption of Autonomous Vehicles (AVs) in developing countries by considering the sustainability dimensions. The barriers are identified by conducting a comprehensive literature review and studying the academic experts' opinions in related industries. By identifying the main barriers to the sustainable adoption of AVs, a synthesized approach of the Rough Best-Worst Method (RBWM) and Interval-Rough Multi-Attributive Border Approximation Area Comparison (IR-MABAC) is utilized for weighting and evaluating each barrier in this context. According to the results of this study, the "inflation rate", "lack of internet connection quality", and "learning challenges and difficulties to use the AVs" are the top challenges and barriers to the AV adoption which need to be considered by policymakers. As the main contribution of this research, we provide efficient insights on a macro policy scale for decision-makers with respect to the main barriers to the implementation of AVs technology. From the AVs literature and to the best of our knowledge, this is the first study of its kind that considers the barriers to the AV technology implementation through the sustainability concept.

Relative sensitivity value (RSV): A metric for measuring input parameter influence in life cycle assessment modeling

Life cycle assessment (LCA) is a commonly used tool to quantify life cycle environmental footprints of products. Uncertainty in LCA modeling, particularly from uncertainty in production practices (represented through input parameter arguments), can lead to incorrect conclusions and hamper decision-making. Characterization of uncertainty through stochastic means and sensitivity analysis is utilized in a small fraction of LCA case studies, and the majority of studies default to scenario analysis due to its lower barrier to implementation and its results are easier to interpret. In this article, we introduce a sensitivity metric, relative sensitivity value (RSV), which allows LCA practitioners to gauge the relative influence of production practices on life cycle impacts in multiple phases and impact categories. Relative sensitivity value bridges the gap between scenario analysis and global sensitivity analysis, and it allows an LCA practitioner to provide an easy-to-interpret metric for quantifying the degree to which incremental changes in production practices influences the life cycle environmental footprint. We present the methodology used to calculate RSV and provide programming code, which can be readily used by an LCA practitioner to calculate RSV for their LCA model. We demonstrate the usage of RSV through a livestock husbandry LCA case study, in which we show how RSV results may be presented and interpreted, and how conclusions regarding production practices may be drawn. Integr Environ Assess Manag 2023;19:547-555. © 2022 SETAC.

Insect-Inspired, Spike-Based, in-Sensor, and Night-Time Collision Detector Based on Atomically Thin and Light-Sensitive Memtransistors

Detecting a potential collision at night is a challenging task owing to the lack of discernible features that can be extracted from the available visual stimuli. To alert the driver or, alternatively, the maneuvering system of an autonomous vehicle, current technologies utilize resource draining and expensive solutions such as light detection and ranging (LiDAR) or image sensors coupled with extensive software running sophisticated algorithms. In contrast, insects perform the same task of collision detection with frugal neural resources. Even though the general architecture of separate sensing and processing modules is the same in insects and in image-sensor-based collision detectors, task-specific obstacle avoidance algorithms allow insects to reap substantial benefits in terms of size and energy. Here, we show that insect-inspired collision detection algorithms, when implemented in conjunction with in-sensor processing and enabled by innovative optoelectronic integrated circuits based on atomically thin and photosensitive memtransistor technology, can greatly simplify collision detection at night. The proposed collision detector eliminates the need for image capture and image processing yet demonstrates timely escape responses for cars on collision courses under various real-life scenarios at night. The collision detector also has a small footprint of ∼40 μm2 and consumes only a few hundred picojoules of energy. We strongly believe that the proposed collision detectors can augment existing sensors necessary for ensuring autonomous vehicular safety.

Automotive Radar Processing With Spiking Neural Networks: Concepts and Challenges

Frequency-modulated continuous wave radar sensors play an essential role for assisted and autonomous driving as they are robust under all weather and light conditions. However, the rising number of transmitters and receivers for obtaining a higher angular resolution increases the cost for digital signal processing. One promising approach for energy-efficient signal processing is the usage of brain-inspired spiking neural networks (SNNs) implemented on neuromorphic hardware. In this article we perform a step-by-step analysis of automotive radar processing and argue how spiking neural networks could replace or complement the conventional processing. We provide SNN examples for two processing steps and evaluate their accuracy and computational efficiency. For radar target detection, an SNN with temporal coding is competitive to the conventional approach at a low compute overhead. Instead, our SNN for target classification achieves an accuracy close to a reference artificial neural network while requiring 200 times less operations. Finally, we discuss the specific requirements and challenges for SNN-based radar processing on neuromorphic hardware. This study proves the general applicability of SNNs for automotive radar processing and sustains the prospect of energy-efficient realizations in automated vehicles.

Shared Automated Electric Vehicle Prospects for Low Carbon Road Transportation in British Columbia, Canada

This study explores the long-term energy use implications of electrification, automation and sharing of road vehicles in British Columbia, Canada. Energy use is first analyzed for the years 1990–2016 for forward forecasting, and hypothetical scenarios ranging from conservative to disruptive, incorporating various effects of road vehicle electrification, sharing and automation, as well as influences of other technology disruptions, such as online shopping and e-learning are presented and used to project the road transportation energy use in B.C. to 2060. Transportation energy use projections are compared to those of the Canadian Energy Regulator (CER). When considering only the effect of vehicle electrification, the scenarios show higher energy savings compared to CER’s scenarios. The combined impact of vehicle electrification and automation leads to decreased energy use to 2060 for all scenarios considered. The energy savings for all scenarios, except for the conservative one, are higher than CER’s projections. When the effects of vehicle electrification, automation and sharing are merged, all scenarios yield energy savings beyond the CER projections. Inclusion of other technology disruptions and the effects of pandemics like COVID-19 reduce transportation demand and provide further energy savings. The BAU scenario given in this study shows energy use decreases compared to 2016 of 26.3%, 49%, 62.24%, 72.1% for the years 2030, 2040, 2050, and 2060 respectively.

The Impact of HVAC on the Development of Autonomous and Electric Vehicle Concepts

Automation and electrification are changing vehicles and mobility. Whereas electrification is mainly changing the powertrain, automation enables the rethinking of the vehicle and its applications. The actual driving range is an important requirement for the design of automated and electric vehicles, especially if they are part of a fleet. To size the battery accordingly, not only the consumption of the powertrain has to be estimated, but also that of the auxiliary users. Heating Ventilation and Air Conditioning (HVAC) is one of the biggest auxiliary consumers. Thus, a variable HVAC model for vehicles with electric powertrain was developed to estimate the consumption depending on vehicle size and weather scenario. After integrating the model into a tool for autonomous and electric vehicle concept development, various vehicle concepts were simulated in different weather scenarios and driving cycles with the HVAC consumption considered for battery sizing. The results indicate that the battery must be resized significantly depending on the weather scenario to achieve the same driving ranges. Furthermore, the percentage of HVAC consumption is in some cases higher than that of the powertrain for urban driving cycles, due to lower average speeds. Thus, the HVAC and its energy demand should especially be considered in the development of autonomous and electric vehicles that are primarily used in cities.

Impact of Penetrations of Connected and Automated Vehicles on Lane Utilization Ratio

Lane Utilization Ratio (LUR), affected by lane selection behavior directly, represents the traffic distribution on different lanes of road section for a single direction. The research on LUR, especially under Penetration Conditions of Connected and Automated Vehicles (PCCAV), is not comprehensive enough. Considering the difficulty in the conduction of real vehicle experiment and data collection under PPCAV, the lane selection model based on phase-field coupling and set pair logic, which considers the full-information of lanes, was used to carry out microscopic traffic simulation. From the analysis of microsimulation results, the basic relationships between Penetration of Connected and Automated Vehicles (PCAV), traffic volume, and Lane-Changing Times, also that between PCAV, traffic volume, and LUR in the basic section of the urban expressway were studied. Moreover, the influence of driving propensity on the effect of PCAVs was also studied. The research results could enrich the traffic flow theory and provide the theoretical basis for traffic management and control.

The impact universe—a framework for prioritizing the public interest in the Internet of Things

The connected technologies of the Internet of Things (IoT) power the world we live in. IoT systems and devices are critical infrastructure—they provide a platform for social interaction, fuel the marketplace, enable the government, and control the home. Their increasing ubiquity and decision-making capabilities have profound implications for society. When humans are empowered by technology and technology learns from experience, a new kind of social contract is needed, one that specifies the roles and rules of engagement for a cyber-social world. In this paper, we describe the “impact universe,” a framework for assessing the impacts and outcomes of potential IoT social controls. Policymakers can use this framework to guide technological innovation so that the design, use, and oversight of IoT products and services advance the public interest. As an example, we develop an impact universe framework that describes the social, economic, and environmental impacts of self-driving cars.

Digital technologies are fundamental to the world we live in. Internet-connected systems and devices are a critical infrastructure: they run power and water systems, they drive cars, planes, and trains, and they have changed how we do business. They provide a platform for social interaction—targeting and modulating a mindboggling set of options.

The increasing ubiquity, decision-making capabilities, and far-reaching impacts of connected technologies have profound implications for individuals and society. They mandate new social controls—policy, regulation, law, standards, recommended practice—that promote the public interest in a rapidly changing environment. Developing effective social controls requires a holistic appraisal of their potential impact on society and the environment.

The “impact universe” is a framework that exposes a broad set of impacts—both quantifiable and qualitative—to assess the benefits and risks of connected systems and devices. Developing an impact universe framework requires a stakeholder to identify benefits and risks in aggregate; it encourages them to focus beyond the single metric valuation that often characterizes the development of social controls for connected systems. It provides a tool for stakeholders to more effectively guide technological innovation, so that the design, development, use, and standardization of connected products and services advances the public interest and promotes social responsibility in a tech-powered world.

LiDAR-driven spiking neural network for collision avoidance in autonomous driving

Facilitated by advances in real-time sensing, low and high-level control, and machine learning, autonomous vehicles draw ever-increasing attention from many branches of knowledge. Neuromorphic (brain-inspired) implementation of robotic control has been shown to outperform conventional control paradigms in terms of energy efficiency, robustness to perturbations, and adaptation to varying conditions. Here we propose LiDAR-driven neuromorphic control of both vehicle's speed and steering. We evaluated and compared neuromorphic PID control and online learning for autonomous vehicle control in static and dynamic environments, finally suggesting proportional learning as a preferred control scheme. We employed biologically plausible basal-ganglia and thalamus neural models for steering and collision-avoidance, finally extending them to support a null controller and a target-reaching optimization, significantly increasing performance.

Life Cycle Greenhouse Gas Emissions for Last-Mile Parcel Delivery by Automated Vehicles and Robots

Increased E-commerce and demand for contactless delivery during the COVID-19 pandemic have fueled interest in robotic package delivery. We evaluate life cycle greenhouse gas (GHG) emissions for automated suburban ground delivery systems consisting of a vehicle (last-mile) and a robot (final-50-feet). Small and large cargo vans (125 and 350 cubic feet; V125 and V350) with an internal combustion engine (ICEV) and battery electric (BEV) powertrains were assessed for three delivery scenarios: (i) conventional, human-driven vehicle with human delivery; (ii) partially automated, human-driven vehicle with robot delivery; and (iii) fully automated, connected automated vehicle (CAV) with robot delivery. The robot's contribution to life cycle GHG emissions is small (2-6%). Compared to the conventional scenario, full automation results in similar GHG emissions for the V350-ICEV but 10% higher for the V125-BEV. Conventional delivery with a V125-BEV provides the lowest GHG emissions, 167 g CO2e/package, while partially automated delivery with a V350-ICEV generates the most at 486 g CO2e/package. Fuel economy and delivery density are key parameters, and electrification of the vehicle and carbon intensity of the electricity have a large impact. CAV power requirements and efficiency benefits largely offset each other, and automation has a moderate impact on life cycle GHG emissions.

High-Resolution Traffic Sensing with Probe Autonomous Vehicles: A Data-Driven Approach

Recent decades have witnessed the breakthrough of autonomous vehicles (AVs), and the sensing capabilities of AVs have been dramatically improved. Various sensors installed on AVs will be collecting massive data and perceiving the surrounding traffic continuously. In fact, a fleet of AVs can serve as floating (or probe) sensors, which can be utilized to infer traffic information while cruising around the roadway networks. Unlike conventional traffic sensing methods relying on fixed location sensors or moving sensors that acquire only the information of their carrying vehicle, this paper leverages data from AVs carrying sensors for not only the information of the AVs, but also the characteristics of the surrounding traffic. A high-resolution data-driven traffic sensing framework is proposed, which estimates the fundamental traffic state characteristics, namely, flow, density and speed in high spatio-temporal resolutions and of each lane on a general road, and it is developed under different levels of AV perception capabilities and for any AV market penetration rate. Experimental results show that the proposed method achieves high accuracy even with a low AV market penetration rate. This study would help policymakers and private sectors (e.g., Waymo) to understand the values of massive data collected by AVs in traffic operation and management.

Mind the gap: Developments in autonomous driving research and the sustainability challenge

Scientific knowledge on autonomous-driving technology is expanding at a faster-than-ever pace. As a result, the likelihood of incurring information overload is particularly notable for researchers, who can struggle to overcome the gap between information processing requirements and information processing capacity. We address this issue by adopting a multi-granulation approach to latent knowledge discovery and synthesis in large-scale research domains. The proposed methodology combines citation-based community detection methods and topic modelling techniques to give a concise but comprehensive overview of how the autonomous vehicle (AV) research field is conceptually structured. Thirteen core thematic areas are extracted and presented by mining the large data-rich environments resulting from 50 years of AV research. The analysis demonstrates that this research field is strongly oriented towards examining the technological developments needed to enable the widespread rollout of AVs, whereas it largely overlooks the wide-ranging sustainability implications of this sociotechnical transition. On account of these findings, we call for a broader engagement of AV researchers with the sustainability concept and we invite them to increase their commitment to conducting systematic investigations into the sustainability of AV deployment. Sustainability research is urgently required to produce an evidence-based understanding of what new sociotechnical arrangements are needed to ensure that the systemic technological change introduced by AV-based transport systems can fulfill societal functions while meeting the urgent need for more sustainable transport solutions.

Rear-end collision warning of connected automated vehicles based on a novel stochastic local multivehicle optimal velocity model

Studying the rear-end early warning methods of connected automated vehicles (CAVs) is useful for issuing early warnings and reducing traffic accidents. Establishing a corresponding driving model according to CAV characteristics is necessary when designing intelligent decision and control systems, especially for the safety speed threshold. However, since traffic systems are stochastic, there are random factors that influence car-following behavior. Therefore, this study proposes a rear-end collision warning method for CAVs based on a stochastic local multivehicle optimal speed (SLMOV) car-following model. First, the SLMOV model is proposed to characterize the car-following behavior of CAVs. Simultaneously, a stability analysis and parameter estimation method are discussed. Second, the safety distance between the CAVs changes with time because the speed of the rear vehicles satisfies the SLMOV model, which is used to calculate the safety probability of rear-end CAV collisions through an analysis of the driving process. The speed threshold is assessed by controlling the rear-end collision probability. Third, next-generation simulation (NGSIM) data are used in an empirical analysis of a rear-end collision warning method on the basis of a parameter estimation of the SLMOV model. The results present the speed thresholds of vehicles under different braking deceleration levels. Finally, the merits and demerits of fixed-speed and variable-speed adjustment time intervals are compared by considering driving safety and comfort as evaluation indexes. A reasonable CAV adjustment time interval of 0.4 s is determined. This result can be used to help develop a vehicle loading rear-end collision warning system.

Review and Meta-Analysis of EVs: Embodied Emissions and Environmental Breakeven

Electric vehicles (EVs) are often considered a potential solution to mitigate greenhouse gas (GHG) emissions originating from personal transport vehicles, but this has also been questioned due to their high production emissions. In this study, we performed an extensive literature review of existing EV life-cycle assessments (LCAs) and a meta-analysis of the studies in the review, extracting life-cycle GHG emission data combined with a standardized methodology for estimating GHG electrical grid intensities across the European Economic Area (EEA), which were used to estimate a set of environmental breakeven points for each EEA country. A Monte Carlo simulation was performed to provide sensitivity analysis. The results of the review suggest a need for greater methodological and data transparency within EV LCA research. The meta-analysis found a subset of countries across the EEA where there is a potential that EVs could lead to greater life-cycle GHG emissions than a comparable diesel counterpart. A policy discussion highlights how EV policies in countries with contrasting GHG electric grid intensities may not reflect the current techno-environmental reality. This paper emphasizes the importance for researchers to accurately depict life-cycle vehicle emissions and the need for EEA countries to enact policies corresponding to their respective contextual conditions to avoid potentially enacting policies that could lead to greater GHG emissions.

Maneuver-Based Objectification of User Comfort Affecting Aspects of Driving Style of Autonomous Vehicle Concepts

Driving maneuvers try to objectify user needs regarding the driving dynamics for a vehicle concept. As autonomous vehicles will not be driven by people, the driving style that merges the individual aspects of driving dynamics, like user comfort, will be part of the vehicle concept itself. New driving maneuvers are, therefore, necessary to objectify the driving style of autonomous vehicle concepts with all its interdependencies relating to the individual aspects. This paper presents a methodology to design such driving maneuvers and includes a pilot study and a user study. As an example, the methodology was applied to the parameters of user comfort and travel time. The driven maneuvers resulted in statistical equations to objectify the interdependencies of these two aspects. Finally, this paper provides an outlook for needed maneuvers in order to tackle the entire driving style with its multidimensional facets.

Assessing the Sustainability Implications of Autonomous Vehicles: Recommendations for Research Community Practice

Autonomous vehicles (AV) are poised to induce disruptive changes, with significant implications for the economy, the environment, and society. This article reviews prior research on AVs and society, and articulates future needs. Research to assess future societal change induced by AVs has grown dramatically in recent years. The critical challenge in assessing the societal implications of AVs is forecasting how consumers and businesses will use them. Researchers are predicting the future use of AVs by consumers through stated preference surveys, finding analogs in current behaviors, utility optimization models, and/or staging empirical “AV-equivalent” experiments. While progress is being made, it is important to recognize that potential behavioral change induced by AVs is massive in scope and that forecasts are difficult to validate. For example, AVs could result in many consumers abandoning private vehicles for ride-share services, vastly increased travel by minors, the elderly and other groups unable to drive, and/or increased recreation and commute miles driven due to increased utility of in-vehicle time. We argue that significantly increased efforts are needed from the AVs and society research community to ensure 1) the important behavioral changes are analyzed and 2) models are explicitly evaluated to characterize and reduce uncertainty.

Assessing the Socioeconomic Impacts of Intelligent Connected Vehicles in China: A Cost–Benefit Analysis

The deployment of intelligent connected vehicles (ICVs) is regarded as a significant solution to improve road safety, transportation management, and energy efficiency. This study assessed the safety, traffic, environmental, and industrial economic benefits of ICV deployment in China under different scenarios. A bottom-up model was established to deal with these impacts within a unified framework, based on the existing theories and literature of ICVs’ cost–benefit analysis, as well as China’s most recent policies and statistics. The results indicate that the total benefits may reach 13.25 to 24.02 trillion renminbi (RMB) in 2050, while a cumulative benefit–cost ratio of 1.15 to 3.06 suggests high cost-effectiveness. However, if the government and industry only focus on their own interests, the break-even point may be delayed by several years. Hence, an effective business model is necessary to enhance public–private cooperation in ICV implementation. Meanwhile, the savings of travel time costs and fleet labor costs play an important part in all socioeconomic impacts. Therefore, the future design of ICVs should pay more attention to the utilization of in-vehicle time and the real substitution for human drivers.

Energy and emissions implications of automated vehicles in the U.S. energy system

Vehicle automation has the potential to drastically transform transportation, with important implications for energy and the environment. There is considerable uncertainty regarding the impact of automation on travel demand and vehicle efficiency. We utilize the MARKet ALlocation (MARKAL) energy system model to examine four previously published scenarios that consider different effects of automation on efficiency and demand. We do not replicate detailed estimation of individual mechanisms but apply key outcomes from prior studies within a broader energy system framework. Our analysis adds insights on fuel switching, upstream impacts, and air emissions. MARKAL dynamically captures interactions between transportation and non-transportation sectors, which is important given that the revolutionary shifts from automation may invalidate static assumptions. Model results suggest that increasing travel demands from automation may boost fuel use and petroleum-based fuel prices, potentially increasing the market penetration of alternative-fuel vehicles. In contrast, dramatic efficiency improvements from automation could drive fuel prices lower, greatly reducing the competitiveness of alternative-fueled vehicles. Furthermore, these shifts could yield positive or negative environmental impacts. Some automation scenarios even resulted in counterintuitive results. For example, if high levels of efficiency improvement drive out alternative-fuel vehicles, such as battery electric and hybrids, a net worsening of air quality relative to the other scenarios could result. We also found system-level dynamics to be key. For example, reductions in liquid fuel prices led to increased consumption, and the resulting increase in air pollutant emissions offset a portion of the potential air quality benefits of automation.

A Review on Energy, Environmental, and Sustainability Implications of Connected and Automated Vehicles

Connected and automated vehicles (CAVs) are poised to reshape transportation and mobility by replacing humans as the driver and service provider. While the primary stated motivation for vehicle automation is to improve safety and convenience of road mobility, this transformation also provides a valuable opportunity to improve vehicle energy efficiency and reduce emissions in the transportation sector. Progress in vehicle efficiency and functionality, however, does not necessarily translate to net positive environmental outcomes. Here, we examine the interactions between CAV technology and the environment at four levels of increasing complexity: vehicle, transportation system, urban system, and society. We find that environmental impacts come from CAV-facilitated transformations at all four levels, rather than from CAV technology directly. We anticipate net positive environmental impacts at the vehicle, transportation system, and urban system levels, but expect greater vehicle utilization and shifts in travel patterns at the society level to offset some of these benefits. Focusing on the vehicle-level improvements associated with CAV technology is likely to yield excessively optimistic estimates of environmental benefits. Future research and policy efforts should strive to clarify the extent and possible synergetic effects from a systems level to envisage and address concerns regarding the short- and long-term sustainable adoption of CAV technology.