The SHRP 2 Database represents the largest current repository of naturalistic driving data in the world. Including over 3,500 drivers (many of them observed over several years), 30 million miles, and encompassing over 2 petabytes of storage, it offers countless possibilities for research in the fields of driver safety, traffic operations, and transportation logistics. Due to the size of the database and its complexity, it can be described as “Big Data” and consequently presents unique challenges in its processing and utilization. These challenges offer opportunities for the development of algorithms that can be used to effectively process even larger data sources in the future. Several novel tools have been developed in order to make the database more accessible. These tools include a dedicated website that can be used to explore the breadth of the dataset (http://insight.shrp2nds.us), map matching tables to provide association between driving data and the surrounding infrastructure-environment, and detailed coding of crashes and near-crashes identified through a series of kinematic triggers executed throughout the data as well as comparable “normal-driving” periods. This paper discusses the development of these tools, their usefulness to future research, and their output. Since the purpose of the dataset is to improve driver safety, this paper also lists several potential future applications of the data. The paper concludes with a detailed description of the process required by researchers to access these data, which includes approval of the research project by an ethics board and the execution of a conditional data use license.

View full abstract

Naturalistic cycling studies can be performed by making instrumented bicycles available to participants, or by having mobile equipment added by the participants themselves to their own bicycles. This paper describes how participants' bicycles can be equipped with a commercially available, small and unobtrusive action camera to gather naturalistic cycling data. Lateral position, swerving and speed were analysed using video and GPS to assess cycling behaviour of older cyclists, and to assess the influence of different types of cycling infrastructure. The applied method gathered insights in possible interventions in a cost efficient way, inconspicuously, and quickly, compared with instrumented bicycles. Also, several bicycles can be instrumented at the same time with as only limitation the number of cameras available.

View full abstract