Air Force Tests Paired Robots for Aircraft Inspection Workflows
An upcoming U.S. Air Force evaluation at the Warner Robins Air Logistics Complex is worth paying attention to, but not just for being another drone in the hangar. What makes it interesting is an attempt to evaluate a complete inspection architecture one aerial robot, one ground robot, one autonomy stack, and one data environment for maintenance.
The Asylon company received a Phase Three contract from Warner Robins to deploy its Multi-modal Autonomous Robotics for Inspection of Aircraft, or MARIA, system at the sustainment site. The solution uses the company’s Guardian small unmanned aircraft system and the DroneDog Q-UGV ground robot. Both are operated using the Range autonomy software from Asylon and connected to the DroneIQ command-and-control platform meant to provide maintenance personnel with a unified interface for mission assignment and mission results analysis.
The system integration is the key engineering accomplishment here. The issue of aircraft inspection is usually viewed as a sensing challenge, but in reality, it is also a workflow problem. Image capturing is just one of many tasks. The bigger challenge is in making aircraft inspections repeatable, selecting the right robot and sending it to the right position for capturing, storing the collected information in a proper format and showing the results in such a way that does not create an additional software layer to check.
At Warner Robins, the task is to evaluate how autonomous systems integration can decrease the time necessary for performing general visual inspections and increase the consistency of repetitive maintenance tasks. The personnel of the Air Force is also going to evaluate if it allows maintaining aircrafts without increasing manpower needs. This is important for big sustainment centers, where the slightest gain in repeatability and review speed will accumulate in time.
The MARIA system is supposed to allow assigning the inspection missions to robotic platforms instead of conducting them via manual walk-around inspections only. Then, the air and ground systems will capture the aircraft imagery and other data, send the results to maintenance personnel for their review. According to the description, this concept should allow performing unattended and repeatable general visual inspections of Department of the Air Force aircrafts and improve inspection quality while decreasing labor-related risks.
This combination of air and ground robots is significant because of peculiar geometry of aircraft surfaces. The upper fuselage and tail parts are better seen by an aerial platform while low-surface areas, landing gear, and side view of aircraft surfaces are better inspected by a ground robot. So, when using both of them under the unified autonomy and command control suggests that Air Force evaluates the coverage and consistency of inspection as a unified task and not two separate robotic tools.
The data environment is also important. DroneIQ is a hub for the outputs of inspections, collecting the imagery, LiDAR point cloud, telemetry, and other operational data. The platform supports both traditional 2D views and digital twin environment. From the viewpoint of maintenance personnel, it could be more significant than robots itself. Unified review environment will simplify documentation, reduce time for analysis of different data sources and make it possible to conduct the repeated inspection of the same aircraft area in a more standard way.
It does not mean that the system replaces maintenance personnel. The current implementation aims at moving MARIA from its development phase to the operational one in maintenance conditions, with the results of inspection still sent to personnel for their review and decision-making. In other words, the autonomy in this case means automation of data collection, mission execution and maintenance workflow, not removing the human factor from the process of aircraft maintenance.
Warner Robins is called the main sustenance facility of Air Force, making it the right place for this kind of testing. A robotic platform could perform well in a demonstration and fail in a real sustainment center environment under the influence of schedule, mixed aircraft geometries and operational requirements. So, testing of such a system in a major sustainment center becomes a crucial step in checking its robustness for routine operations.
According to Asylon, this award expands the Air Force portfolio of the company from autonomous security operations. Anthony McCarty, the company’s Senior Vice President of Government, stated that at Warner Robins, they will test MARIA in an operational maintenance environment before any deployment in sustainment programs of the defense.
So, the message is clear. If this test succeeds, the biggest change would not be the usage of a flying robot or a robotic dog separately. It would be validation of a whole inspection architecture, including sensing, autonomy, data management and review of maintenance personnel as one sustenance solution. For U.S. military aviation, it could become a practical robotics revolution without science fiction promises.
By Jonathan Barrett — Editor for AMI’s future mobility and autonomous systems section with two decades of coverage of robotics, e-mobility, drone-vehicle convergence, and transport mechanical systems.
