Vehicle simulators are often built from scratch, using well-established software frameworks with custom vehicle models. Since the development of a simulator is a very time consuming process, the purpose of this thesis paper was to explore a different alternative by evaluating the open-source vehicle system ChaosVehicle (CV) in Unreal Engine 5. The goal was to build a simulator using CV and assess its performance in imitating certain aspects of a three-wheeled electric vehicle, including acceleration, coasting and braking behaviors in a straight line.The simulator was set up according to the data provided by the vehicle manufacturer, OMotion. By comparing the simulations with real world driving data, it was determined that the simulator achieved a relative error below 16% in imitating the coasting behavior of the real vehicle. Further, with regards to the brake performance, the simulator has managed to achieve a relative error of below 15% when not accounting for velocities below 10 km/h. Due to the noise present in the real world driving data, it was inconclusive as to whether the simulator had accurately predicted the acceleration behavior of the real vehicle. After a preliminary analysis of the plugin, it was discussed whether CV was as extensible as hypothesized. While the framework is open-source, it was difficult to fully grasp the system due to incomplete documentation of the source code and the user interface. Further work in mapping how the physics model in CV is integrated to the game engine would shed some light on this matter.
| Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:mau-68603 |
| Date | January 2024 |
| Creators | Kuan, Dick, Zettel, Lucas Svensson |
| Publisher | Malmö universitet, Institutionen för datavetenskap och medieteknik (DVMT) |
| Source Sets | DiVA Archive at Upsalla University |
| Language | English |
| Detected Language | English |
| Type | Student thesis, info:eu-repo/semantics/bachelorThesis, text |
| Format | application/pdf |
| Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.002 seconds