How do long combination vehicles perform in real traffic? A study using Naturalistic Driving Data
利用自然驾驶数据评估两种长组合车辆（A-double和DuoCAT）在换道、环岛、路口转弯和急弯四种场景下的稳定性和操控性，并分析驾驶员认知负荷。
This paper evaluates the performance of two different types of long combination vehicles (A-double and DuoCAT) using naturalistic driving data across four scenarios: lane changes, manoeuvring through roundabouts, turning in intersections, and negotiating tight curves. Four different performance-based standards measures are used to assess the stability and tracking performance of the vehicles: rearward amplification, high-speed transient offtracking, low-speed swept path, and high-speed steady-state offtracking. Also, the steering reversal rate metric is employed to estimate the cognitive workload of the drivers in low-speed scenarios. In the majority of the identified cases of the four scenarios, both combination types have a good performance. The A-double shows slightly better stability in high-speed lane changes, while the DuoCAT has slightly better manoeuvrability at low-speed scenarios like roundabouts and intersections. • Performance of long combination vehicles (A-double and DuoCAT) evaluated using naturalistic driving data. • Four scenarios analysed: lane changes, roundabouts, intersections, and tight curves. • Performance-based standards and steering reversal rate used to assess stability, tracking, and driver workload. • A-double showed slightly better high-speed stability while a slightly better low-speed manoeuvrability observed with DuoCAT.