How to use zed cameras to help move an autonomous toy car to a destination while avoiding obstacles

Hello all, I have a zed 2i camera on a toy car (with an nvidia jetson nano) and I am setting up an obstacle course for the car to drive autonomously with a person on the other end as the destination for the car. I am using another camera with a different jetson to detect the person and send the persons location to the car. I am trying to integrate positional tracking, relocalization, object detection, and spatial mapping to complete this project. I am using python to run everything and have been looking through the zed git hub page, but I am a very inexperienced coder and am struggling to integrate together everything the zed cameras have to offer. How do I integrate different aspects of zed using python to complete my task? Thanks for all the help, Grant.

Hello. I don’t know Python, I’m dabbling with Unity and C#, and while your project sounds really awesome, it also sounds quite large to be able to combine all of those things. Breaking the project up in to manageable chunks may make it easier. A good starting point might be to try and find an object detection example in Python for the Jetson, and just get that to work. If it detects an object, have the jetson blink an LED using a GPIO pin.

You know what, forget all of that. I think the thing to do is to first get a platform to work in. The platform will pull it all together for you. The Unity 3D (C#) development platform is fun, and the ROS 2 (Python) platform is very good. Start a project in one of those, add the ZED plugin, follow an example for spatial mapping, and enjoy seeing the results rendered on the screen. That’ll get you going :slight_smile: