Ballistic weapons, whether launched from a cannon, long-range missile, or aircraft, go through a deceleration phase before reaching a “steady-state flight” state, at which point the missile can automatically detect and identify its target. To ensure accuracy, extensive testing must be conducted early in development to measure the rotation rate and sweep angle of the orbit (collectively known as flight attitude).
Measurement of these critical parameters is done by on-board or off-board sensors. On-board sensors such as accelerometers, gyroscopes, geomagnetic sensors, and solar sensors are typically mounted inside the bullet body during testing, while off-board sensors such as optical measuring devices, radar, and global positioning satellite systems are typically placed outside the bullet body. Regardless of the type of sensor, bullet tests are performed in a vertical wind tunnel to simulate real flight conditions.
Optical measurement devices are increasingly being used to test bullets. Thanks to high-speed video, cameras combined with computer vision software can be used to smoothly observe the steady rotation of a bullet at hundreds of revolutions per second. One drawback of optical measurement is that although this technology is generally more accurate at measuring rotational speed than other types of sensors, it lacks precision in measuring the scan angle.
Simple and flexible optical method
Researchers from the School of Aeronautics at Northwestern Polytechnical University and Xi’an Institute of Modern Control Technology have designed a new, extremely simple, yet versatile, optical measurement system that can accurately measure both sweep angle and rotation rate during stable flight. Whereas traditional optical methods use multiple cameras, the new system uses just one 3-megapixel EoSens 3CXP camera from Mikrotron.
The team tested the proposed solution in a university laboratory equipped with a vertical wind tunnel. Wind speeds in the tunnel range from 5 to 50 m/s (meters per second) to reproduce the projectile trajectory. Each projectile is suspended in the air by a soft rope that is threaded through a pulley and rotates around a plumb line. Depending on the type of projectile, the rotation speed is maintained at 4 to 30 revolutions per second. The researchers placed a Mikrotron camera in a fixed position with a resolution of 1280 x 1024 pixels, a focal length of 35 mm, and a frame rate of 1000 frames per second, utilizing a CoaXPress high-speed interface. To ensure sufficient illumination, an LED light source is also used.
Once the rotating object reached a steady state, the microtron camera recorded a total of 5,000 images over a period of 5 seconds. The rotation speed is determined by tracking the shape and coordinates of the rotating object’s features using a tracking or deep learning algorithm. Furthermore, a mathematical model is presented to calculate the scan angle of a rotating object in steady state, which makes it possible to measure the scan angle from images recorded by the camera only from a fixed position. The algorithm efficiently captures and analyzes the rotational motion of the rotating object, accurately measuring both the sweep angle and the rotation speed. The low measurement error demonstrates the reliability and robustness of the algorithm in accurately quantifying these parameters.
