The computer keyboard is one of the most common, reliable, accessible, and effective tools used for humanÀmachine interfacing and information exchange. Although keyboards have been used for hundreds of years for advancing human civilization, studying human behavior by keystroke dynamics using smart keyboards remains a great challenge. Here we report a self-powered, non-mechanical-punching keyboard enabled by contact electrification between human fingers and keys, which converts mechanical stimuli applied to the keyboard into local electronic signals without applying an external power. The intelligent keyboard (IKB) can not only sensitively trigger a wireless alarm system once gentle finger tapping occurs but also trace and record typed content by detecting both the dynamic time intervals between and during the inputting of letters and the force used for each typing action. Such features hold promise for its use as a smart security system that can realize detection, alert, recording, and identification. Moreover, the IKB is able to identify personal characteristics from different individuals, assisted by the behavioral biometric of keystroke dynamics. Furthermore, the IKB can effectively harness typing motions for electricity to charge commercial electronics at arbitrary typing speeds greater than 100 characters per min. Given the above features, the IKB can be potentially applied not only to self-powered electronics but also to artificial intelligence, cyber security, and computer or network access control.

A B S T R A C TElectric vehicles (EVs) have great potential to reduce transportation-related fossil fuel consumption as well as pollutant and greenhouse gas (GHG) emissions, due to their use of renewable electricity as the sole energy source. Therefore, the wide-spread deployment of EVs is regarded as an attractive means to mitigate the environmental problems (e.g., air pollution and climate change) resulting from transportation activities. Government agencies are trying to promote EV deployment by allocating considerable funding as well as promulgating supportive policies. However, the mass adoption of EVs is still impeded by the limited charging infrastructure and allelectric-range (AER). All these lead to a critical research topic: the EV energy consumption analysis and estimation under real-world traffic conditions, which is fundamental to various types of EV-centred applications that aim at improving the EV energy efficiency and extending the AER. For example, eco-routing systems for EVs rely on accurate link-level energy consumption estimation to calculate the EV energy consumption costs of the different route options. In this work, to obtain an accurate link-level energy consumption estimation model for EVs, the energy consumption under real-world traffic congestion is decomposed based on two proposed impact factors: positive kinetic energy (PKE) and negative kinetic energy (NKE). Upon this decomposition, a data-driven model is built to estimate EV energy consumption on each roadway link considering real-world traffic conditions. Finally, the model performance is evaluated by comparing with the performance of baseline model adapted from existing models. The results show that the proposed EV link-level energy consumption estimation model outperforms the existing models in terms of accuracy, implying that it is quite promising in various on-board EV applications.

E nergy management is a critical technology in plug-in hybrid-electric vehicles (PHEVs) for maximizing efficiency, fuel economy, and range, as well as reducing pollutant emissions. At the same time, deep reinforcement learning (DRL) has become an effective and important methodology to formulate model-free and realtime energy-management strategies for HEVs and PHEVs. In this article, we describe the energy-management issues of HEVs/PHEVs and summarize a variety of potential DRL applications for onboard energy management. We also discuss the prospects for various DRL approaches in the energy-management field.

Plug-in hybrid electric vehicles (PHEVs) show great promise in reducing transportation-related fossil fuel consumption and greenhouse gas emissions. Designing an efficient energy management system (EMS) for PHEVs to achieve better fuel economy has been an active research topic for decades. Most of the advanced systems rely either on a priori knowledge of future driving conditions to achieve the optimal but not real-time solution (e.g., using a dynamic programming strategy) or on only current driving situations to achieve a real-time but nonoptimal solution (e.g., rule-based strategy). This paper proposes a reinforcement learning–based real-time EMS for PHEVs to address the trade-off between real-time performance and optimal energy savings. The proposed model can optimize the power-split control in real time while learning the optimal decisions from historical driving cycles. A case study on a real-world commute trip shows that about a 12% fuel saving can be achieved without considering charging opportunities; further, an 8% fuel saving can be achieved when charging opportunities are considered, compared with the standard binary mode control strategy.