SimViz is a Unity package for auto-pilot system developers to create their simulation environment easily in Unity.
You can acquire the package here:
When I worked in Unity, I helped developed several SimViz projects for customers or potential customers to meet their specific needs.
General Case:
While more customers want us to create a workable simulation environment for them. Usually to be their parking lot or a piece of testing area in real world. In that case, we need to take sample photos first.
Then create correspondant testing environment with basic shapes.
Then upgrade textures and materials.
After environment was textured and rendering was upgraded, we need to set up the AI move path and test, which was basically list of positions and rotations for the vehicles to pass one by one.
The last part was to adjust the parameters of LiDar / Radar / RGB-D camera. Particularly, is the "turbulance" difference and latency simulation for each specific environments, such as windy, snowy, heavy rain, etc.
Particular Case:
1. RTCP Streaming
For some customer, they want SimViz system to output RTCP streaming, because their algorithm was based on RTCP Camera.
After researching, I prototyped Unity with RTCP streaming. But I also figured that RTCP is a cached protocol, that it has to cache several pieces of pictures then together push to clients. By which means the latency is inevitable. I also suggested to the customers about the regular UDP plan. For more info about streaming, you could check my another blog to get more details.
2. HDRP integration.
Some customers needs better quality, and I tried upgraded with high definition rendering pipeline.
But Lidar and radar simulation is not supported in HDRP, as they are implemented based on geometry shaders.
An alternative way to work around is to also introduce Visual Effect Graph (VEG), which is compatible to customized rendering pipeline.
All you need to do is to calculate the distance to each objects in the environment and then implement the detecting faces of the visual effects, setting the particles to hit on the face.