How Hyper built a 1M-accurate indoor GPS
andrewhart.me
How Hyper built a 1M-accurate indoor GPS
1–10 of 75 posts
Re: How Hyper built a 1M-accurate indoor GPS
#2Whoa, 1m is really impressive! I've played around with stuff like https://github.com/dmsl/anyplace for https://www.emfcamp.org/ back in the day, but that was around 5m at best. Bet you a robotics startup will scoop it up!
Re: How Hyper built a 1M-accurate indoor GPS
#3How does your system handle the massive variance in sensor quality (accelerometer, gyro, WiFi radio) between a high-end iPhone and a budget Android device? Does the 1m accuracy hold up across the board, or does it degrade gracefully? Getting this right seems critical for scaling to a 'billion people'
Re: How Hyper built a 1M-accurate indoor GPS
#4Congrats on the launch, that looks dope! I'm curious, will this be able to run on an embedded robotics hardware?
Re: How Hyper built a 1M-accurate indoor GPS
#5Congrats on the launch, that looks dope! I'm curious, will this be able to run on an embedded robotics hardware?
Yes - one of the limitations with mobile is engaging the camera/quality of the SLAM. With a robot, they're already using SLAM with strong tracking, and controlling the hardware stack means no device limitations (uncalibrated sensors, limited WiFi pings etc.)
Re: How Hyper built a 1M-accurate indoor GPS
#6Whoa, 1m is really impressive! I've played around with stuff like https://github.com/dmsl/anyplace for https://www.emfcamp.org/ back in the day, but that was around 5m at best. Bet you a robotics startup will scoop it up!
Thanks. Very interested in robotics so that'd be an ideal home, given the SLAM + sensors tech stack.
Re: How Hyper built a 1M-accurate indoor GPS
#7How does your system handle the massive variance in sensor quality (accelerometer, gyro, WiFi radio) between a high-end iPhone and a budget Android device? Does the 1m accuracy hold up across the board, or does it degrade gracefully? Getting this right seems critical for scaling to a 'billion people'
All of the videos shown on the speed-run of our technology is on 4 year-old Android devices, such as Samsung S21, Pixel 6 etc.. We always test and gather statistics on older devices to fairly represent what's available, rather than latest-and-greatest.
Re: How Hyper built a 1M-accurate indoor GPS
#8Very, very impressive, if (intentionally?) a little vague on how you got to 1m accuracy given you WiFi only gives you 3m.
I guess you must use the constraints on the directions a person can walk imposed by the shelves and other structures to give you orientation of the accelerometers. Which in turn means the person doing the ground truth mapping must walk down every aisle, and into every gap. That's not so difficult if your staff are doing it, but I bet you have trouble training the store staff gathering that data to do it well.
Best of British.
Re: How Hyper built a 1M-accurate indoor GPS
#9Very, very impressive, if (intentionally?) a little vague on how you got to 1m accuracy given you WiFi only gives you 3m. I guess you must use the constraints on the directions a person can walk imposed by the shelves and other structures to give you orientation of the accelerometers. Which in turn means the person doing the ground truth mapping must walk down every aisle, and into every gap. That's not so difficult…
Thanks for the feedback. I wanted to keep it balanced to be accessible but also insightful.
To answer your point: we have the digital map, can use that to understand obstacles etc in the space. In some of those larger stores you see in the visual, we typically survey the entire store within 2-3 hours, it's low-effort work, not a blocker.
Re: How Hyper built a 1M-accurate indoor GPS
#10Why do you need a human to do the initial mapping ? why can't you use a Roomba or a smaller hardware for this task