When developing algorithms for autonomous vehicles it is important to test several different scenarios many times. New and untested algorithms are prone to make errors which results in accidents. It is therefore preferred to use a simulation environment instead. Sensors used to determine the vehicle’s position must then be modelled. This thesis answers the question whether adding sensor simulation to an existing simulation platform (AGXUnity) is viable, or if using other existing options is preferred. To reach the goal, a function-based sensor is developed and its accuracy tested. Its performance is determined by simulation in a standard scene. Tests showed that the sensor had acceptable accuracy and performance. The conclusion is that AGXUnity is a viable platform for sensor simulation.
| Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:umu-158017 |
| Date | January 2019 |
| Creators | Kallin, Niklas |
| Publisher | Umeå universitet, Institutionen för datavetenskap |
| 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 |
| Relation | UMNAD ; 1977 |
Page generated in 0.0022 seconds