The development of computationally efficient and validated single-track vehicle-rider models has traditionally required handcrafted one-off models. Here we introduce BRiM, a software package that facilitates building these models in a modular fashion while retaining access to the mathematical elements for handcrafted modeling when desired. We demonstrate the flexibility of the software by constructing the Carvallo-Whipple bicycle model with different numerical parameters representing different bicycles, modifying it with a front fork suspension travel model, and extending it with moving rider arms driven by joint torques at the elbows. Using these models we solve a lane-change optimal control problem for six different model variations which solve in mere seconds on a modern laptop. Our tool enables flexible and rapid modeling of single-track vehicle-rider models that give precise results at high computational efficiency.

Despite the publication of many bicycle models, there is yet to exist a common framework for building and extending bicycle-rider models that can be easily shared between researchers. To fill this gap, we have developed BRiM, a modular and extensible open-source framework for creating Bicycle-Rider Models. It uses an established bicycle model, like the Carvallo-Whipple model, which can be extended using components from BRiM's extensive library, or user-defined custom subclasses. It leverages the open-source Python package SymPy, a computer algebra system, to compute the equations of motion. This results in symbolic equations of motion, which, after code generation, can be used to simulate and optimise the model. The effectiveness of BRiM is demonstrated by solving a trajectory tracking problem using a direct collocation algorithm.