In the topics quiz focusing on laserscan readings to avoid the wall, the traditional method is to use only 1 ray of readings (e.g ranges[360] for the front reading).
However, I find this solution not robust enough. Instead, I thought about creating a detection cone based on a range of readings. Thus, I was wondering if there were norms to define these detection cones (arbitrary dividing the laserscan reading zone in 3, or 1 big for the front and 2 small for left / right).
Is there any way here to apply such a thing, and if so, to which resources should I focus?