Creating a safe workplace environment is an investment. Manufacturers can prevent workplace injuries by kickstarting a culture of safety, according to Voxel, a San-Francisco-based start-up that uses artificial intelligence (AI) technology to try to prevent accidents before they happen.
Alex Senemar, CEO, Voxel AI, explains in the accompanying video that his team built generalized AI models that can deploy out of the box. The AI safety solution is an add-on to the security cameras used in manufacturing and industrial warehouse facilities for monitoring operational workflow, safety and security.
“We look at environmental hazards, whether that be a spill on the floor that can cause a slip, trip and fall or a blocked exit, and we look at hazards around ergonomics,” said Senemar. “People are making repetitive movements and motions that could lead to injury over time. You can think of poor lifting or overreaching as examples.”
Voxel’s platform is set up via an edge computing device that connects directly to existing security camera systems. Video feeds are then directed to the Cloud, where footage can be analyzed. Senemar explained that red-flag incidents—including hazards, risky behaviors and operational inefficiencies—are reported to the client, who can duly assign it to a supervisor for addressing with staff. The system can be up and running within 48 hours at customer sites.
“We look at the leading indicators of risk and leading indicators of safety and injuries that take place in the workplace and we try to identify them via the existing security cameras, using the artificial intelligence and computer vision models that we have developed,” explained Senemar, who previously co-founded Sherbit, an AI-powered remote health monitoring system for hospitals (acquired in 2018).
At this point, the AI models have seen hundreds of thousands of hours of video, and Senemar said the technology has been able to train on different versions and iterations of these events that connect directly to the camera.
Voxel’s founding team is made up of a talent pool of former colleagues who had largely cut their teeth while working in the self-driving car space. Based on that experience, Voxel was able to apply key learnings that have evolved in computer vision and AI since it launched in 2020. Today, the start-up is expanding what was gleaned from the automotive industry and exploring how the models can generalize and scale across different industrial environments.
“When you look at these large companies, they have quite a large footprint of industrial areas where injuries can take place,” said Senemar.