Modelling, predicting and analysing driver behaviours are essential to advanced driver assistance systems (ADAS) and the comprehensive understanding of complex driving scenarios. Recently, with the development of deep learning (DL), numerous driver behaviour learning (DBL) methods have been proposed and applied in connected vehicles (CV) and intelligent transportation systems (ITS). This study provides a review of DBL, which mainly focuses on typical applications in CV and ITS. First, a comprehensive review of the state-of-the-art DBL is presented. Next, Given the constantly changing nature of real driving scenarios, most existing learning-based models may suffer from the so-called “catastrophic forgetting,” which refers to their inability to perform well in previously learned scenarios after acquiring new ones. As a solution to the aforementioned issue, this paper presents a framework for continual driver behaviour learning (CDBL) by leveraging continual learning technology. The proposed CDBL framework is demonstrated to outperform existing methods in behaviour prediction through a case study. Finally, future works, potential challenges and emerging trends in this area are highlighted.

The shedding of pathogens by infected humans enables the use of sewage monitoring to conduct wastewater-based epidemiology (WBE). Although most WBE studies use data from large sewage treatment plants, timely data from smaller catchments are needed for targeted public health action. Traditional sampling methods, like autosamplers or grab sampling, are not conducive to quick ad hoc deployments and high-resolution monitoring at these smaller scales. This study develops and validates a cheap and easily deployable passive sampler unit, made from readily available consumables, with relevance to the COVID-19 pandemic but with broader use for WBE. We provide the first evidence that passive samplers can be used to detect SARS-CoV-2 in wastewater from populations with low prevalence of active COVID-19 infections (0.034 to 0.34 per 10,000), demonstrating their ability for early detection of infections at three different scales (lot, suburb, and city). A side by side evaluation of passive samplers (n = 245) and traditionally collected wastewater samples (n = 183) verified that the passive samplers were sensitive at detecting SARS-CoV-2 in wastewater. On all 33 days where we directly compared traditional and passive sampling techniques, at least one passive sampler was positive when the average SARS-CoV-2 concentration in the wastewater equaled or exceeded the quantification limit of 1.8 gene copies per mL (n = 7). Moreover, on 13 occasions where wastewater SARS-CoV-2 concentrations were less than 1.8 gene copies per mL, one or more passive samplers were positive. Finally, there was a statistically significant (p < 0.001) positive relationship between the concentrations of SARS-CoV-2 in wastewater and the levels found on the passive samplers, indicating that with further evaluation, these devices could yield semi-quantitative results in the future. Passive samplers have the potential for wide use in WBE with attractive feasibility attributes of cost, ease of deployment at small-scale locations, and continuous sampling of the wastewater. Further research will focus on the optimization of laboratory methods including elution and extraction and continued parallel deployment and evaluations in a variety of settings to inform optimal use in wastewater surveillance.

A greedy heuristic-based algorithm was put forward to obtain the approximate optimal solution for the DUDC(discrete unit disk cover) problem within polynomial time. Firstly, discrete cells were generated to represent the 2D plane. Then, alternative sets which can cover certain amount of target points were built. A greedy algorithm was articulated to find a minimal combination of the alternative sets, and full coverage of all target points was realized. The minimal covering circles were calculated considering the position of each point in the sets. The center of the minimal covering circles was selected as the location. Based on a case study, the effectiveness of the proposed algorithm was validated. The factors of influence the performance of the algorithm were also discussed and ratio of time complexity and approximation was analyzed.