Explore the vast range of titles available.

One critical bottleneck that impedes the development and deployment of autonomous vehicles is the prohibitively high economic and time costs required to validate their safety in a naturalistic driving environment, owing to the rarity of safety-critical events 1 . Here we report the development of an intelligent testing environment, where artificial-intelligence-based background agents are trained to validate the safety performances of autonomous vehicles in an accelerated mode, without loss of unbiasedness. From naturalistic driving data, the background agents learn what adversarial manoeuvre to execute through a dense deep-reinforcement-learning (D2RL) approach, in which Markov decision processes are edited by removing non-safety-critical states and reconnecting critical ones so that the information in the training data is densified. D2RL enables neural networks to learn from densified information with safety-critical events and achieves tasks that are intractable for traditional deep-reinforcement-learning approaches. We demonstrate the effectiveness of our approach by testing a highly automated vehicle in both highway and urban test tracks with an augmented-reality environment, combining simulated background vehicles with physical road infrastructure and a real autonomous test vehicle. Our results show that the D2RL-trained agents can accelerate the evaluation process by multiple orders of magnitude (10 3 to 10 5 times faster). In addition, D2RL will enable accelerated testing and training with other safety-critical autonomous systems. An intelligent environment has been developed for testing the safety performance of autonomous vehicles and its effectiveness has been demonstrated for highway and urban test tracks in an augmented-reality environment.

Driving intelligence tests are critical to the development and deployment of autonomous vehicles. The prevailing approach tests autonomous vehicles in life-like simulations of the naturalistic driving environment. However, due to the high dimensionality of the environment and the rareness of safety-critical events, hundreds of millions of miles would be required to demonstrate the safety performance of autonomous vehicles, which is severely inefficient. We discover that sparse but adversarial adjustments to the naturalistic driving environment, resulting in the naturalistic and adversarial driving environment, can significantly reduce the required test miles without loss of evaluation unbiasedness. By training the background vehicles to learn when to execute what adversarial maneuver, the proposed environment becomes an intelligent environment for driving intelligence testing. We demonstrate the effectiveness of the proposed environment in a highway-driving simulation. Comparing with the naturalistic driving environment, the proposed environment can accelerate the evaluation process by multiple orders of magnitude. Tests for autonomous vehicles are usually made in the naturalistic driving environment where safety-critical scenarios are rare. Feng et al. propose a testing approach combining naturalistic and adversarial environment which allows to accelerate testing process and detect dangerous driving events.

Centralized water treatment has dominated in developed urban areas over the past century, although increasing challenges with this model demand a shift to a more decentralized approach wherein advanced oxidation processes (AOPs) can be appealing treatment options. Efforts to overcome the fundamental obstacles that have thus far limited the practical use of traditional AOPs, such as reducing their chemical and energy input demands, target the utilization of heterogeneous catalysts. Specifically, recent advances in nanotechnology have stimulated extensive research investigating engineered nanomaterial (ENM) applications to AOPs. In this Perspective, we critically evaluate previously studied ENM catalysts and the next-generation treatment technologies they seek to enable. Opportunities for improvement exist at the intersection of materials science and treatment process engineering, as future research should aim to enhance catalyst properties while considering the unique roadblocks to practical ENM implementation in water treatment.

Population and development megatrends will drive growth in cement production, which is already one of the most challenging-to-mitigate sources of CO 2 emissions. However, availabilities of conventional secondary cementitious materials (CMs) like fly ash are declining. Here, we present detailed generation rates of secondary CMs worldwide between 2002 and 2018, showing the potential for 3.5 Gt to be generated in 2018. Maximal substitution of Portland cement clinker with these materials could have avoided up to 1.3 Gt CO 2 -eq. emissions (~44% of cement production and ~2.8% of anthropogenic CO 2 -eq. emissions) in 2018. We also show that nearly all of the highest cement producing nations can locally generate and use secondary CMs to substitute up to 50% domestic Portland cement clinker, with many countries able to potentially substitute 100% Portland cement clinker. Our results highlight the importance of pursuing regionally optimized CM mix designs and systemic approaches to decarbonizing the global CMs cycle. In this paper we report the maximum potential for cement substitution with secondary materials to reduce CO2 emissions globally (1.3 Gt CO2-eq. in 2018) and on a country-by-country basis.

Proper disposal of industrial brine has been a critical environmental challenge. Zero liquid discharge (ZLD) brine treatment holds great promise to the brine disposal, but its application is limited by the intensive energy consumption of its crystallization process. Here we propose a new strategy that employs an advanced solar crystallizer coupled with a salt crystallization inhibitor to eliminate highly concentrated waste brine. The rationally designed solar crystallizer exhibited a high water evaporation rate of 2.42 kg m −2 h −1 under one sun illumination when treating real concentrated seawater reverse osmosis (SWRO) brine (21.6 wt%). The solar crystallizer array showed an even higher water evaporation rate of 48.0 kg m −2 per day in the outdoor field test, suggesting a great potential for practical application. The solar crystallizer design and the salt crystallization inhibition strategy proposed and confirmed in this work provide a low-cost and sustainable solution for industrial brine disposal with ZLD. Proper disposal of industrial brine remains a critical environmental challenge. Here, the authors devise a solar crystallizer and propose a salt crystallization inhibition strategy, which together provide a low-cost and sustainable solution for industrial brine disposal with zero liquid discharge.

Traffic in an urban network becomes congested once there is a critical number of vehicles in the network. To improve traffic operations, develop new congestion mitigation strategies, and reduce negative traffic externalities, understanding the basic laws governing the network’s critical number of vehicles and the network’s traffic capacity is necessary. However, until now, a holistic understanding of this critical point and an empirical quantification of its driving factors has been missing. Here we show with billions of vehicle observations from more than 40 cities, how road and bus network topology explains around 90% of the empirically observed critical point variation, making it therefore predictable. Importantly, we find a sublinear relationship between network size and critical accumulation emphasizing decreasing marginal returns of infrastructure investment. As transportation networks are the lifeline of our cities, our findings have profound implications on how to build and operate our cities more efficiently.

The sensing of gravity has emerged as a tool in geophysics applications such as engineering and climate research 1 – 3 , including the monitoring of temporal variations in aquifers 4 and geodesy 5 . However, it is impractical to use gravity cartography to resolve metre-scale underground features because of the long measurement times needed for the removal of vibrational noise 6 . Here we overcome this limitation by realizing a practical quantum gravity gradient sensor. Our design suppresses the effects of micro-seismic and laser noise, thermal and magnetic field variations, and instrument tilt. The instrument achieves a statistical uncertainty of 20 E (1 E = 10 −9  s −2 ) and is used to perform a 0.5-metre-spatial-resolution survey across an 8.5-metre-long line, detecting a 2-metre tunnel with a signal-to-noise ratio of 8. Using a Bayesian inference method, we determine the centre to ±0.19 metres horizontally and the centre depth as (1.89 −0.59/+2.3) metres. The removal of vibrational noise enables improvements in instrument performance to directly translate into reduced measurement time in mapping. The sensor parameters are compatible with applications in mapping aquifers and evaluating impacts on the water table 7 , archaeology 8 – 11 , determination of soil properties 12 and water content 13 , and reducing the risk of unforeseen ground conditions in the construction of critical energy, transport and utilities infrastructure 14 , providing a new window into the underground. A study reports a quantum gravity gradient sensor with a design that eliminates the need for long measurement times, and demonstrates the detection of an underground tunnel in an urban environment.

From stadium covers to solar sails, we rely on deployability for the design of large-scale structures that can quickly compress to a fraction of their size 1 – 4 . Historically, two main strategies have been used to design deployable systems. The first and most frequently used approach involves mechanisms comprising interconnected bar elements, which can synchronously expand and retract 5 – 7 , occasionally locking in place through bistable elements 8 , 9 . The second strategy makes use of inflatable membranes that morph into target shapes by means of a single pressure input 10 – 12 . Neither strategy, however, can be readily used to provide an enclosed domain that is able to lock in place after deployment: the integration of a protective covering in linkage-based constructions is challenging and pneumatic systems require a constant applied pressure to keep their expanded shape 13 – 15 . Here we draw inspiration from origami—the Japanese art of paper folding—to design rigid-walled deployable structures that are multistable and inflatable. Guided by geometric analyses and experiments, we create a library of bistable origami shapes that can be deployed through a single fluidic pressure input. We then combine these units to build functional structures at the metre scale, such as arches and emergency shelters, providing a direct route for building large-scale inflatable systems that lock in place after deployment and offer a robust enclosure through their stiff faces. Origami-inspired multistable structures that can be inflated from flat to three dimensions have been designed; a library of foldable shapes is created and then combined to build metre-scale functional structures.

Streamflow ( Q flow ) prediction is one of the essential steps for the reliable and robust water resources planning and management. It is highly vital for hydropower operation, agricultural planning, and flood control. In this study, the convolution neural network (CNN) and Long-Short-term Memory network (LSTM) are combined to make a new integrated model called CNN-LSTM to predict the hourly Q flow (short-term) at Brisbane River and Teewah Creek, Australia. The CNN layers were used to extract the features of Q flow time-series, while the LSTM networks use these features from CNN for Q flow time series prediction. The proposed CNN-LSTM model is benchmarked against the standalone model CNN, LSTM, and Deep Neural Network models and several conventional artificial intelligence (AI) models. Q flow prediction is conducted for different time intervals with the length of 1-Week, 2-Weeks, 4-Weeks, and 9-Months, respectively. With the help of different performance metrics and graphical analysis visualization, the experimental results reveal that with small residual error between the actual and predicted Q flow , the CNN-LSTM model outperforms all the benchmarked conventional AI models as well as ensemble models for all the time intervals. With 84% of Q flow prediction error below the range of 0.05 m 3  s −1 , CNN-LSTM demonstrates a better performance compared to 80% and 66% for LSTM and DNN, respectively. In summary, the results reveal that the proposed CNN-LSTM model based on the novel framework yields more accurate predictions. Thus, CNN-LSTM has significant practical value in Q flow prediction.

The heating and cooling energy consumption of buildings accounts for about 15% of national total energy consumption in the United States. In response to this challenge, many promising technologies with minimum carbon footprint have been proposed. However, most of the approaches are static and monofunctional, which can only reduce building energy consumption in certain conditions and climate zones. Here, we demonstrate a dual-mode device with electrostatically-controlled thermal contact conductance, which can achieve up to 71.6 W/m 2 of cooling power density and up to 643.4 W/m 2 of heating power density (over 93% of solar energy utilized) because of the suppression of thermal contact resistance and the engineering of surface morphology and optical property. Building energy simulation shows our dual-mode device, if widely deployed in the United States, can save 19.2% heating and cooling energy, which is 1.7 times higher than cooling-only and 2.2 times higher than heating-only approaches. Future zero-energy buildings require smart and dynamic utilization of renewable energy for efficient indoor temperature control. Here the authors show that the dual-mode device enables building envelopes to switch between solar heating and radiative cooling to save HVAC energy for all seasons and all climate zones.