Why the Air Force wants to install lidar on robotic dogs

Lidar on robotic dogs

What would it take to rebuild an Air Force base after a direct hit? In 2018, Hurricane Michael tore through Tyndall Air Force Base in Florida, damaging both aircraft and facilities. Rebuilding the site and protecting against future disasters is an ongoing challenge and is viewed by the Air Force as a training exercise. To that end, the company is experimenting with how laser sensors on drones and robotic dogs can show damage before a person has to enter a collapsing building.

“Imagine being able to see the components of a potentially dangerous situation in real time in 3D and in great detail without having to look around you,” says Brian Goddin of the Public Affairs Center at the Air Force Installation and Mission Support Center. in a video produced by the army. .

As he speaks, the video lights up the interior of the garage. The laser-generated vision is surreal, almost supernatural, with objects not seen as complete shapes, but as reflected lines adjacent to each other in space. The imaging tool is lidar, which stands for light and range detection, and while the results are a bit strange, they are sharp enough to demonstrate roof damage in a garage. Construction work around the facility and a large armored car parked at the end are also visible.

Installing lidar on drones and ground robots gives the military the ability to map the interior of a building using a machine. With this lidar data transmitted to computers in the command center or even to the tablet of an operator sitting outside the building, a person can see what the robot sees and steer it accordingly. (In the civilian world, lidar sensors are commonly used in autonomous cars as one of the tools to allow vehicles to detect the world around them.)

Goddin’s presentation, posted online on December 9, 2021, shows a Spot-mounted lidar, a dog-shaped robot from Boston Dynamics. Ghost Robotics’ Q-UGVs, also dog-shaped and equipped with sensors, were used to patrol the perimeter of Tyndall AFB, making Spot the second breed (or brand) of robot dogs for the base.

While all of this rendering in Tyndale takes place after Hurricane Michael, creating a virtual 3D model of the buildings as they are could help with future renovations. Such a virtual model is a useful tool for routine maintenance and repair, and provides a record of past status in the event of a recurring disaster.

These methods may also allow for better ex post facto investigation of the failure. By comparing lidar scans of a sunken or wrecked ship with pre-launch scans and surviving aircraft returning from combat, the Air Force was able to discover the best way to make a more durable ship. Scanning a crashed plane with lidar also lets rescuers and recovery teams know if and how they should act to rescue pilots and passengers, suggested Javier Rodríguez, a Tyndall-based technician.

“Lidar is the gold standard because we can get information that we couldn’t get with simple images,” said Sean Cloud, Air Force Rapid Airfield Damage Assessment Program Manager, describing in the same video a use case to repair. runway stripes. after an attack. “We can calculate the amount of material we need for repairs based on the funnels and the amount of backfill.”

Counting the amount of concrete needed to repair a pothole is not the most interesting use for robots and lasers. Yet it is this totally unappealing kind of routine that allows runways to remain operational and keep planes in the sky ..