CarSim with non-ideal sensors
A newbie to Gazebo asks:
We are creating a car-sim (don't care about the car mechanics, joints, axle's, tires) but we do want realistic sensors: LIDAR, Doppler Radar, Stereo Vision.
What I mean by this is we want to create a 3D world (urban-scape) where we can simulate rain, snow, and other effects, simulate realistic effects for the sensors. This is not the simple noise or camera distortion issues (the tutorials cover that).
An example would be is that we find our stereo vision sensor (mounted on the car) are great for objects moving together with our car, position reporting is erratic at distance (exponential fall-off) as well as horizontally moving objects get mis-classified.
Basic question: can we create sensor plug-ins that can both do these probabilistic effects, based on programmatic changes to the world (SDF) model?
Do these plug-ins need to be in C++ or can they also be in JavaScript+NodeJS
