While many legacy components of the Internet of Things (IoT) are static and fixed to their deployed position (e.g. surveillance cameras, environment sensors), some of the most important pieces are increasingly evolving to be ones which move either on their own accord (e.g. vehicles, robots, drones) or with us (e.g. mobile devices, wearables). This produces a rapidly expansible ecosystem with dynamic and non-deterministic workloads. This fluidity moving devices introduce stretches IoT management platforms’ capabilities, which now require extension with advanced automation to handle the requirements of seamless service delivery in an energy-efficient manner. In November 2024, a methodology to measure connected device emissions was announced by the Carbon Trust in collaboration with Amazon, Microsoft, Samsung and Sky – the energy use of connected devices is currently equivalent to the total energy use of France. In parallel, there are service concerns, with the Evening Standard asking in August 2024 why the UK’s phone signals are so bad. These are therefore pressing areas of investigation with much potential for positive contributions to be made. Autonomous IoT, where intelligent algorithms take over IoT management, is a promising approach for a Sustainable IoT. The previous catalyst considered a very simple scenario of one “moving thing”, a drone, hopping between different edge nodes causing the need of “follow-the-thing” type of IoT service and network orchestration. Additionally, we considered sustainability by applying simple techniques for energy savings. In this proposal, we focus on increasing the scale of the problem. How can we handle the main pain points of service and network orchestration and sustainability planning when the volume of “moving things” increases significantly e.g. thousands of drones, or thousands of “moving things” of different nature e.g. drones, autonomous vehicles etc which share the same infrastructure.