There were some developments not (yet) reported here. For example, the user cognitive load actively controlling a robot and drone simultaneously was onerous!
As mentioned on X though it’s perhaps time to take stock of numerous experiments, developments & test results to create some kind of Minimum Viable Product (MVP).
That is, it’s tempting to keep adding yet more bells and whistles yet this can result in losing sight of the original objectives.
For instance, those (very) early days of simulating swarms of bots searching for “mines” using NetLogo still have a certain inextricable resonance/allure.
Those original notions of emergence, complexity and even earlier ideas of ‘simple’ machines rather than elaborate MCU controlled all singing, all dancing devices need to be revisited.