A new Views & Comments piece published in Engineering unpacks complex carbon trade-offs embedded within large-scale autonomous vehicle (AV) deployment, weighing vehicle-level emission reductions against overlooked environmental burdens tied to road infrastructure and energy grids. Authored by Xiaobo Qu from Tsinghua University alongside Daniel Sperling and Hui Li from the University of California, Davis, the commentary grounds its analysis in recent industry partnerships and integrated vehicle-road system research to offer a holistic assessment of automated mobility’s climate footprint.

The report opens by documenting fast-moving commercialisation milestones from early 2025, including strategic tie-ups between Waymo and Toyota, Pony.ai with Uber and Tencent Cloud, which signal maturing AV technology spanning five automation tiers up to fully driverless Level 5 operation. Core vehicle-side decarbonisation mechanisms outlined include vehicle-to-vehicle communication that smooths traffic flow, gradual acceleration and deceleration programming, and automated platooning that cuts aerodynamic drag for trailing heavy trucks. The paper also notes natural alignment between autonomous systems and electric vehicles (EVs), whose simplified powertrains ease control integration while enabling access to renewable electricity supplies, further lowering transport greenhouse gas output. Shared autonomous vehicles (SAVs) add supplementary efficiency by lifting fleet utilisation rates, cutting private car ownership demand, minimising empty deadhead travel, and freeing urban land previously reserved for parking for green space and public amenities, while expanding accessible mobility for elderly, underage and disabled groups.

The commentary devotes substantial focus to understudied offsetting carbon pressures originating from road infrastructure. Automated platooning creates concentrated, channelised wheel loading that accelerates pavement degradation, raising maintenance frequency and associated infrastructure emissions. Combined with heavier battery packs in electric heavy-duty trucks that amplify surface stress, regular repair cycles drive up lifecycle carbon output for road networks. Grid strain presents another trade-off: mass simultaneous charging of electric AV fleets creates new electricity peak loads requiring power network upgrades, though connected AV fleets can deliver balancing benefits via vehicle-to-grid smart charging timed to off-peak hours. The authors also flag divergent travel outcomes based on ownership models; privately owned AVs risk steep rises in vehicle miles travelled as occupants conduct work or leisure tasks during trips, erasing efficiency gains, while pooled shared fleets deliver far lower per-passenger energy use.

Integrated vehicle-road system modelling cited in the text records modest average cumulative emission reductions from autonomous driving, with larger savings achievable through multi-rider shared services. To mitigate conflicting environmental impacts, the paper lays out coordinated technical and policy pathways covering vehicle lightweight chassis design, durable pavement materials, embedded road sensing for predictive maintenance, vehicle-road-cloud digital twin systems, and regulatory frameworks prioritising shared autonomous mobility via fleet subsidies, dynamic occupancy-linked road pricing and urban space reallocation. The commentary closes by calling for cross-sector collaboration to advance lifecycle environmental impact research, align vehicle engineering with infrastructure resilience, and design targeted policies that steer AV adoption toward equitable, low-carbon transport outcomes.

The paper “Carbon Trade-Offs of Autonomous Vehicles in Transportation Systems,” is authored by Xiaobo Qu, Daniel Sperling, Hui Li. Full text of the open access paper: https://doi.org/10.1016/j.eng.2025.10.036. For more information about Engineering, visit the website at https://www.sciencedirect.com/journal/engineering.