A new angle detection method based on the compound parabolic collector (CPC): Design and simulation

Abstract

Imaging optics play a crucial role in the angle measurement system. However, the field of view (FOV) is inversely proportional to the focal length in the optical imaging system, meaning that the detection distance and FOV cannot be considered simultaneously. This paper presents a non-imaging angle detection system incorporating an optical model using a compound paraboloid collector (CPC) and data processing based on a backpropagation (BP) neural network. This method has a detection capability of 200 km, with the FOV of azimuth and pitch angles being 60°. The angle detection resolution is 1°, while the angle error is less than 0.1°. A detector array absorbs the far-field light reflected by the CPC. Changing the pitch and azimuth angles establishes the nonlinear relationship between the number of rays each detector receives and the angle. Then, a BP neural network is used to train the nonlinear relationship into a linear one. Based on the light flux data on the detector array, the simulation results show that the system can effectively predict the angle.