The future is now. Self-driving ag machines are rolling across farm fields. But while it might seem like recent breakthroughs in artificial intelligence have suddenly enabled automation in agriculture, its development has been a long time coming. “I’ve been saying, ‘It’s next year,’ for 25 years,” said Mel Torrie, founder and CEO of Autonomous Solutions Inc. (ASI). Autonomous machines work by using sensors to gather data about their surroundings such as obstacles, terrain and weather. Software then analyzes this data and tells the machine what to do next: turn left, speed up, stop or adjust its operation. The foundations of machine learning dates back to World War II, where it was developed to predict aircraft trajectories and weather trends. Innovations like GPU chips accelerate processing, enabling edge computing, or the ability to process information directly within devices rather than through cloud technology. So, why now? Torrie said the conditions are now right for automation. “We have this convergence of labor challenges and governments being willing to bring some standards. It is time, and so we are now pedal to the metal,” he said. He speaks from experience, having developed ag tech machine automation for more than 26 years. Torrie launched ASI in 2000 after John Deere discovered a research paper he wrote while in a master’s program in electrical engineering at Utah State University. “They asked us to do some initial pilot projects at the university lab where I was working. They wanted us to prove that we could safely navigate a tractor around a 3-year-old, so we put a mannequin on a remote control toy and played chicken with the tractor,” he recalled. Since those early experiments, ASI has helped develop the back-end automation technology used by major ag brands today. Developing automation for ag tech is challenging because