BehaviorTree.CPP VS carla

Have a look at Behavior Trees, especially the BehaviorTreeCpp library https://www.behaviortree.dev The method is simple to understand, scales better than State Machines and if you implement the actions well they’re inherently modular. Nav2 uses them, and also BostonDynamics in their Spot SDK.

Unrelated to javascript, but in the robotics industry there is a trend towards Behaviour Trees. https://www.behaviortree.dev/ is a C++ library that was originally designed for controlling robots via ROS, but it appears to be decoupled from the ROS ecosystem so it could be used for other projects.

There is a basic UI available for editing the tree https://github.com/BehaviorTree/Groot

It sounds like you're looking for something like the CARLA simulator.

If you are looking for a different simulator, I would suggest using (Carla)[https://carla.org/] with ROS bridge and it also has an inbuilt support for OSM which worked flawlessly (you have to install it from source to get the OSM plugin).

(3) I would suggest you use CARLA or TORCS for self-driving cars in RL as they are common test beds.

carla env

Link to carla

There are a few ways you could go with this. A fun recent trend has been to make simulators in the unreal engine for photorealistic training. If you wanted to spend your whole project on the simulator part you could make your own environment, but I highly recommend using an open-source sim package. If you don't care too much about photorealism, you could use gazebo just fine. I've also made small visual worlds in blender, then simulated with RVIZ. Here's a good one with a focus on aerial robotics: https://theairlab.org/tartanair-dataset/ You could also give CARLA a spin for autonomous vehicles: https://carla.org/