Ansible Motion, manufacturer of driver-in-the-loop (DIL) simulators, and the University of Cambridge’s Engineering Department (CUED) are working together in support of a PhD study into human driver behavior, funded by Toyota Motor Europe.
The industry-academia partnership has set out to compare novice and expert drivers’ steering behavior during extreme vehicle maneuvers, in order to reveal distinct learning patterns and control styles.
The simulator
The Delta S3 DIL simulator that will be used in the study features a patented six-degree-of-freedom Stratiform motion system and wrap-around projection graphics, which creates an immersive environment for collecting precise data on driver responses. The simulator has ultra-low latency and large excursion capabilities to ensure realistic dynamics. With an open and modular software architecture, the simulator can be integrated with sophisticated vehicle and subsystem physics models, hardware-in-the-loop (HIL) test benches and other essential vehicle development toolchains.
The simulator features advanced motion and vision and audio immersion capabilities, and a fully integrated real cabin environment for virtual vehicle testing and development.
The study
For over 25 years, researchers at CUED have examined driver-vehicle dynamics to enhance vehicle design and safety.
In this latest study, initial experiments were based on existing experimental track data from an instrumented vehicle where drivers had repeatedly performed identical obstacle avoidance maneuvers. The data identified differences in steering and learning strategies between drivers. However, the repeated maneuvers used for the existing data limited the learning on the evolution of a driver’s steering strategy.
To combat this, a new virtual test-driving experiment was devised in which 20 test subjects steered a vehicle along randomly curving paths using Ansible Motion’s dynamic Delta S3 DIL simulator. Software from Ansible Motion’s sister company, rFPro, assisted in generating early results that confirmed differing learning rates between drivers, validating the virtual approach.
The final analysis comparing the real-world data and Ansible Motion’s simulator data with the theoretical model will be completed in late 2025, offering further insights for future vehicle design and driver assistance systems.
David Cole, professor of mechanical engineering at CUED, said, “Cambridge University Engineering Department has been researching driver-vehicle dynamics for over a quarter of a century, aiming to improve vehicle design, performance and safety. Our most recent work, with funding from Toyota Motor Europe, is analyzing how novice and expert drivers learn steering control.
“In order to collect experimental data in a safe and consistent environment, we made the decision to use a driving simulator with immersive motion and graphics. We are very grateful to Ansible Motion for their generous in-kind contribution to the research, by providing access to their simulator software, hardware and technical support. PhD researcher Harry Fieldhouse took 20 test subjects to Ansible Motion’s R&D Centre and performed the experiment, which ran according to plan.”
Dan Clark, managing director of Ansible Motion, added, “Studies such as this bridge the gap between academia and industry, so we’re proud to support Cambridge University Engineering Department’s ground-breaking research into human driver behavior. This investigation underscores the role of vehicle simulation in shaping the future of automotive engineering and development.”
In related news, the first tenant at the new MotorSport Vision Innovation Park in Snetterton, Norfolk, will be Ansible Motion with its simulation center. Click here to read the full story