According to Chris Nielsen — founder and CEO of Levatas, the company that developed this smart Spot — the robodoggies have an AI-based cognitive inspection platform capable of accomplishing tasks like autonomously detecting and reading analog gauges, identifying thermal anomalies, checking for unauthorized people, looking for signs of corrosion and equipment damage, scanning for unsafe conditions in the workplace (like spills and leaks), and a lot of other valuable industrial jobs.
In an email interview, Nielsen says these abilities have enabled large factory sites to use these robots in automated reconnaissance missions instead of installing thousands of expensive sensors everywhere. Each of these robodog missions have long tasks lists detailing what they need to check out during every walk. These tasks result in large datasets of observations that get fed into a database. “Only technical people can handle these. At the end of each mission, the robots capture a ton of data. There’s no simple way to query all of it on demand,” he says.
That’s where ChatGPT comes in. Nielsen says his team has created a much more streamlined way for Spot and its human controllers to communicate using natural language. Humans can talk to Spot using normal language commands, but perhaps more useful is Spot’s new ability to instantly parse tons of information and use those insights to answer previously unanswerable questions. “For example, we can cross-reference information from different [reconnaissance] missions for the first time without pre-programing that capability,” Nielsen says. “We give ChatGPT the raw data and instructions on interpreting it, and it answers the customer’s requests.”
Being able to direct, query, and obtain previously hidden insights about an industrial site’s status is a big deal for these companies. The ChatGPT integration now means anyone can talk to Spot, not just people with technical expertise and robotic training. “Customers can query anything they want, and only the information they need, using plain English,” Nielsen says.
Adding a ChatGPT brain to these machines means they can now provide an extra layer of safety, too. Rather than depending on a single remote operator, any human in the facility will be able to safely and quickly pause or alter the robot’s autonomous mission, without requiring technical training. They just need to speak to it.
Nielsen says his team is currently investigating how to use GPT-4 to solve complex tasks, not just follow commands, he says. “The big leap will happen when we get large language models as capable as GPT-4 running locally on devices instead of online,” he says, which limits where you can use these robots, as many of these large facilities don’t have Internet connection throughout the site.
Once they get these new complex brains running independently inside the robots’ guts, they will be able to let a few of them loose inside any facility, give them a core mission, and have them all figure out how to best perform it without any technical training. “These Spot units will be able to explore facilities, discover a variety of equipment and conditions, and then report back with safety, security, and maintenance recommendations for the human teams to address,” he says.
And then, only then, you will be able to worry about somebody installing poisonous darts and human zappers in the robodogs, Charlie Brooker.
©Fast Company 2023 Mansueto Ventures, LLC., Distributed by Tribune Content Agency, LLC.
