How precise do you need to be? The trick with this question (unless I’m misunderstanding) is that the answer isn’t simply yes or no; you have to define how you want to deal with partial visibility. Potential options range from a small number of raycasts (if you’re not too worried about partial visibility and the obstacles you care about are all pickable) to precomputed or periodically updated visibility maps (if your environment is relatively constrained and stable such that it makes sense to preprocess it) to shaders evaluating the rendered depth/scene pixel by pixel (if you need things very precise and have the resources – both dev and compute – to achieve this).
Picking a few important points on your target (for example, the center and/or corners of your ground plane in your Playground) and raycasting at them is almost certainly the easiest way to get off the ground, though it’s not really efficient and could be imprecise. Will that work? If not, can you tell us a little more about your use case?