Mechanical Simulation Corporation has released the 2019.1 versions of the vehicle dynamics simulation tools CarSim and TruckSim, which are well known for providing high-fidelity math models that reproduce physical test results, yet run fast enough on the computer to support real-time simulation with hardware in the loop. In addition to modeling the physics of vehicle dynamics, CarSim and TruckSim include built-in controllers to provide steering, throttle, and braking controls as needed to simulate traditional vehicle testing and new scenarios needed by the automotive industry to evaluate advanced driver assistance systems (ADAS) and autonomous vehicles (AV).
According to David Hall, Senior Development and Consulting Engineer, “With the increasing emphasis on ADAS and AV, simulations need to represent limit conditions and control challenges that include low speeds and driving in reverse.”
In past versions, closed-loop steering was provided by a linear quadratic regulator (LQR) with optimal control and preview from ten points on a target path. A new controller was introduced in 2019.1 that uses a single target point and one parameter for preview. It uses a simple geometric solution that remains valid well beyond the linear range, and inherently compensates for non-linear tire properties. Dependence on a single target point means the target can readily be imported from external software with users only needing to provide X- and Y-coordinates of a target point.
Both steering controllers (single point and LQR) have been extended to function when driving in reverse. Simply providing a negative target speed and directing the transmission to be in reverse invoke the reverse-travel logic. As with driving forward, the single point control can receive its target from external sources or calculate it from a defined internal path.
In addition to steering by setting the steering wheel angle, steering by torque is now supported by both controllers. A steering wheel torque is applied to achieve the target steer angle calculated by the controller.
Many ADAS and AV controls have logic based on control of vehicle acceleration rather than speed. Acceleration and braking can now be controlled with target accelerations.
“With the new control options,” Hall notes, “engineers developing ADAS and AV capabilities like park assist or developing path identification strategies have more tools to evaluate their systems.” Controlling target accelerations and path following in reverse extend the range of real-world challenges that can be studied. These features enable interaction with third-party software in more convenient ways.