so im just confused. is it not possible to incorporate computervision with the wingman core? i was just wondering if it even could be possibe. it may help with the context awareness issue for the AI. if it can gain context from a constant source of images/video it could help with decision making so that if someone does say something like "evasive manuvers" it could get the proper trajectories and information to execute the commands in sequence? idk im probably just rambling nonsense at this point but it had me thinking