Pycrash

Pycrash is a 2D mathematical model for simulating vehicle motion and vehicular impacts using a variety of approaches based on fundamental physics and published accident reconstruction techniques. In addition to fundamental applications, Pycrash has accident reconstruction specific modules for obtaining and processing commonly used data

Installation (Requires Python > 3.6)

pip install pycrash PyPi

Planar Vehicle Motion

1. Single vehicle motion based on driver inputs (brake, throttle, steer)
2. Multi-vehicle motion and multiple impacts per vehicle

Impact Simulation

Single Degree of Freedom Model

• basic 2D collision with user defined stiffness values and braking effects
• vehicle specific data is published for most vehicles by the National Highway Traffic Safety Administration
• use published stiffness data such as Bonugli et al. or Lee et al. as examples

Impulse Momentum Planar Collision

• model based on conservation of momentum incorporating inter-vehicular sliding developed by Carpenter et al.

Sideswipe (in progress)

• similar to previous model developed and validated by Funk et al.
• improved to allow for driver inputs and flexible approach angles and contact locations

Validation

• validation data are provided within the Github repo
• Initial publication in 2021 SAE
• Validation against real-world crashes 2026 SAE

Modular Design

• Model calculations divided into modules to allow for future improvements
• Vehicle motion incorporates a modular linear tire model to determine slip condition based on weight shift from longitudinal and lateral acceleration.

Collaboration

• Contributors are welcome - create a branch to develop a new capability or improve the current model
• There is a list of projects in the report for future applications that I would like to pursue
• Create an issue if you identify a problem or to make suggestions for future improvements
• Additional validation data are always welcome

Recommendations

• Download Python here and install required packages with pip from PyPi
• See Github repo for tutorials, Jupyter Notebooks and validation data
• The validation and demonstration of Pycrash is best viewed using Jupyter notebooks