At this year’s World Artificial Intelligence Conference (WAIC) and World Robotics Conference (WRC), the focus of demonstrations of embodied intelligence has shifted towards practical applications. Robots are now being used in real-world scenarios such as logistics, industry, and home applications. The industry consensus is that embodied intelligence exhibits a clear “long-tail effect”: the first 60% to 80% of progress occurs quickly, but the last 20% of implementation challenges mainly stem from the complexity of real-world scenarios and the challenges of cloud infrastructure. From demonstrating capabilities to actually completing tasks, robots need to handle system issues such as unstable networks, unknown situations, data transmission processing, and manual remote control in case of task failures. Currently, robots still do not fully possess autonomous capabilities, and when they exceed their autonomous boundaries, reliance on cloud support becomes crucial.
# The last 20% of embodied intelligence implementation is hidden in “the cloud”
At this year’s WAIC and WRC, the trend of embodied intelligence exhibits has changed: it no longer focuses on showing off technology but turns to practical applications. Robots are beginning to enter real-world scenarios such as logistics, industry, and home use.
After visiting companies like Lingchu Intelligence and Aio Intelligence and discussing with several cloud experts, a common understanding gradually emerged: tasks in real-world scenarios are much more complex than demo scripts.
Cloud experts pointed out that embodied intelligence has a clear “long-tail effect” – the first 60%, 70%, and 80% of progress happens quickly, but the last 20% requires a lot of time to be filled. And this last 20% lies precisely in the complexity of real-world scenarios and the challenges of cloud infrastructure.
Between demonstrating capabilities and completing tasks, there is a whole system in between. This system has to deal with various “unexpected” situations in the real world: unstable networks, robots encountering uncharted circumstances, data needing to be processed after transmission, and manual remote control when tasks fail.
## Let robots “migrate to the cloud”
Today’s robots are still far from complete autonomy. When robots exceed their autonomous capabilities…