Kingsley Adjenughwure

arXiv2024

Safety-Oriented Calibration and Evaluation of the Intelligent Driver Model

Many car-following models like the Intelligent Driver Model (IDM) incorporate important aspects of safety in their definitions, such as collision-free driving and keeping safe distances, implying that drivers are safety conscious when driving. Despite their safety-oriented nature, when calibrating and evaluating these models, the main objective of most studies is to find model parameters that minimize the error in observed measurements like spacing and speed while studies specifically focused on calibrating and evaluating unobserved safe behavior captured by the parameters of the model are scarce. Most studies on calibration and evaluation of the IDM do not check if the observed driving behavior (i.e. spacing) are within the model estimated unobserved safety thresholds (i.e. desired safety spacing) or what parameters are important for safety. This limits their application for safety driven traffic simulations. To fill this gap, this paper first proposes a simple metric to evaluate driver compliance with the safety thresholds of the IDM model. Specifically, we evaluate driver compliance to their desired safety spacing, speed and safe time gap. Next, a method to enforce compliance to the safety threshold during model calibration is proposed. The proposed compliance metric and the calibration approach is tested using Dutch highway trajectory data obtained from a driving simulator experiment and two drones. The results show that compliance to the IDM safety threshold greatly depends on braking capability with a median compliance between 38% and 90% of driving time, indicating that drivers can only partially follow the IDM safety threshold in reality.

cs.RO↗

arXiv2023

PRISMA: A Novel Approach for Deriving Probabilistic Surrogate Safety Measures for Risk Evaluation

Surrogate Safety Measures (SSMs) are used to express road safety in terms of the safety risk in traffic conflicts. Typically, SSMs rely on assumptions regarding the future evolution of traffic participant trajectories to generate a measure of risk, restricting their applicability to scenarios where these assumptions are valid. In response to this limitation, we present the novel Probabilistic RISk Measure derivAtion (PRISMA) method. The objective of the PRISMA method is to derive SSMs that can be used to calculate in real time the probability of a specific event (e.g., a crash). The PRISMA method adopts a data-driven approach to predict the possible future traffic participant trajectories, thereby reducing the reliance on specific assumptions regarding these trajectories. Since the PRISMA is not bound to specific assumptions, the PRISMA method offers the ability to derive multiple SSMs for various scenarios. The occurrence probability of the specified event is based on simulations and combined with a regression model, this enables our derived SSMs to make real-time risk estimations. To illustrate the PRISMA method, an SSM is derived for risk evaluation during longitudinal traffic interactions. Since there is no known method to objectively estimate risk from first principles, i.e., there is no known risk ground truth, it is very difficult, if not impossible, to objectively compare the relative merits of two SSMs. Instead, we provide a method for benchmarking our derived SSM with respect to expected risk trends. The application of the benchmarking illustrates that the SSM matches the expected risk trends. Whereas the derived SSM shows the potential of the PRISMA method, future work involves applying the approach for other types of traffic conflicts, such as lateral traffic conflicts or interactions with vulnerable road users.

cs.RO↗