Event-based sensors detect only changes in the brightness across a field of view, with each pixel producing an asynchronous stream of spatial-temporal events data, rather than recording frames of overall illumination of a scene like a traditional frame-based sensor. The location, timestamp, and polarity (positive or negative change) of the recorded events provides low bandwidth data that is useful in real-time sensing of changes in the field of view. These features result in advantages such as high temporal resolution, low latency, high dynamic range, and low power consumption. The use of these sensors has been explored in areas such as object tracking, robotics, and space situational awareness. We are proposing a new application of the event-based sensor in wavefront sensing for adaptive optics. At the Advanced Instrumentation and Technology Centre (AITC) in the Australian National University (ANU), a test bench has been built to test this type of sensor. This study will present the results of this laboratory testing to characterise an event-based sensor’s capabilities to detect tip-tilt aberrations in different conditions, such as background illumination levels. In particular, the event-based sensor’s immunity to a constant background illumination was investigated by introducing an external light source into the scene and recording its performance for different signal strengths. We will present the results of the bench testing, the error budget for tip-tilt sensing using these event-based detectors, and a characterisation of the response of event-based sensors to different illumination conditions.
|