I recently learned that a Santa Fe high school is testing autonomous robots with multiple cameras to monitor the surrounding areas of the school to assist school security team. The autonomous robots learn the routes of the school, capture videos of the surroundings, and provide the footage to school's security team.
This is a great use case for robotics.
- The problem that this robotic system intends to solve is high impact.
- Routes near schools are bounded and not very hard to learn.
- The training data for the computer tasks here are abundant.
- Other data (e.g., behavioral and contextual) can also be added and improve algorithm performance and reduce bias.
- The robots are able to provide higher availability and ground surveillance than what security guards can provide at a given time.
- Imagine a world where insights from the robots can automatically and in near real-time trigger actions of school buildings (e.g., doors automatically lock) and safety measures.
Of course, to implement this at scale, we need to preserve the data privacy of students and teachers while maximizing the value of these robotic systems. There are ways to do this through what data get captured, retained, and analyzed.
The compute, AI/computer vision, and orchestration stack is mature to serve this use case at scale today. I can see this becoming a part of the future of school infrastructure.