Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain
the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in
Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles
and JavaScript.
As autonomous vehicles redefine transportation, addressing challenges from refining sensor technologies to societal implications becomes imperative. This Collection provides a platform to delve into the multiple facets of autonomous driving and traffic systems, fostering a deeper understanding of the challenges and opportunities that lie down the road. Beyond technical developments in perception, sensing and detection systems, navigation algorithms, and control mechanisms, we also welcome submissions focusing on the integration of autonomous vehicles into existing traffic systems, the dynamics of human-machine interaction, and the socio-economic impacts during the coexistence of human and autonomous drivers.
Finding an optimal shape for transport networks, represented as multilayer structures, is a challenging problem. The authors propose analytical and computational frameworks to analyze sharp transitions from symmetric to asymmetric shapes in optimal networks, that can be applied for planning and development of improved multimodal transportation systems within a city.
Without relying on any infrastructure-based vehicle detectors, the authors present a scalable traffic signal re-timing system that uses a small percentage of connected vehicle trajectories as the only input. Real-world tests demonstrate that the system decreases both delays and number of stops.
Ambiguity in human-oriented traffic laws poses a significant challenge to the regulation of self-driving vehicles. Here, the authors present a trigger-based hierarchical online compliance monitor for self-assessment of self-driving vehicles using ambiguous compliance threshold selection principles.
Heavy traffic jams are difficult to predict due to the complexity of traffic dynamics. The authors propose a framework to unveil identifiable early signals and predict the eventual outcome of traffic bottlenecks, which may be useful for designing effective methods preventing traffic jams.
Device imperfections are a major challenge that limit the potential of analogue neural networks. Nanyang Ye and colleagues propose a training-time noise injection approach to improve their robustness without hardware modifications, which comes with theoretical guarantees.
Simulation of naturalistic driving environment for autonomous vehicle development is challenging due to its complexity and high dimensionality. The authors develop a deep learning-based framework to model driving behavior including safety-critical events for improved training of autonomous vehicles.
Over-reliance on automation in transportation systems is known to cause accidents. To address this, here, Tomohiro Nakade and colleagues describe a collaborative strategy for autonomous steering, in which the driver can take over from the automation without its full deactivation.