Case Study: RoboTire Ushers in the Future of Automated Automotive Care

Summary

Story

Victor Darolfi is the CEO of RoboTire and has a long track record in engineering manufacturing and working to improve process efficiency at companies such as Lockheed Martin. While having the tires changed on his car some years back, he drew on his experience to see how greater efficiency and better safety could be brought to the common task of changing tires. He recognized that the current state of robotics, AI, and 3D machine vision was at a point where a new solution was possible. He also saw that if executed correctly, it would change and improve the industry in a truly disruptive way. 

Challenges

The challenges were plentiful but possible to solve. The operation of removing and replacing tires seems a straightforward task, but there are a lot of variables that an automated system needs to be able to handle. The system faces an almost endless range of vehicles, wheels, fastening nuts, and tire locations. It would need a mobile set of eyes on the robot itself to account for all the combinations it would encounter, meaning a robot-mounted 3D camera or sensor. It would also need millimeter-level accuracy to locate wheel nuts and remove them without risking damage to the rims or nuts. It then has to come in at the proper angle, within a fraction of a degree to remove the wheel from the vehicle. Not only must the system do the entire removal, but then it also has to put the new tires back on the vehicle, accurate to a millimeter or submillimeter location, within a fraction of a degree in placement in angle. For this, accuracy is king. Last but not least, the system would need to be highly adaptable and autonomous with minimum operator intervention.  

“Because of the challenges this task presents, we needed a 3D sensor that could maintain a high degree of calibration fidelity and could re-calibrate simply and efficiently. Zivid Two provides this," said Benjamin Wilson,
operations senior manager of Robotire.

Solution

RoboTire chose the Zivid Two industrial 3D camera as their vision solution. This centered around the need for extreme accuracy and an ability to handle a vast variety of scenarios. It necessitated a 3D machine vision system that was designed to excel whilst mounted on the robot’s arm and always produce top-notch results.  As the tire-changing scenarios the system had to cope with had so many variables, they had to have very detailed high-quality point clouds. In addition, the wheel nuts presented one of the biggest challenges in industrial machine vision, how to capture detailed high-quality point clouds from shiny metallic parts. The Zivid Two 3D camera had the high dynamic range (HDR) capabilities to meet this challenge with a dynamic range of over 100 dB.

Robotire's wheel-changing system with a Zivid Two 3D camera

Another factor was the calibration stability of the 3D camera. There is a lot of temperature variation with bay doors opening and closing temperature variations and vibrations would be commonplace. These are all things that will make a 3D sensor system drift if it isn’t designed to cope with such conditions. Fortunately, Zivid 3D cameras are called industrial for a good reason as Zivid design, build, and test each one for calibration resilience, and stability. RoboTire uses advanced AI as a part of its part recognition system for locating the wheel nuts and that AI relies heavily on the accuracy and consistency of the images and point clouds it receives from Zivid Two.  

Result

The system from RoboTire is proving itself with great success on its first deployments. Operators can now focus on other valuable tasks and leave the heavy lifting to the automated tire-changing system from RoboTire. The typical time it takes for the system to change four tires on a car is under 20 minutes, which is very comparable to how fast a human could do it. But perhaps most importantly, this system provides a level of accuracy and repeatability that humans cannot match. Damage to wheel nuts and cross-threading on the vehicle's hubs is a common problem with human operation. When this happens the damage can involve a brand-new wheel unit which is a considerable unexpected expense and inconvenience for the consumer. The RoboTire system practically eliminates this possibility, leaving happy customers and tire shops alike.