Autonomous marine craft navigation: On the study of radar obstacle detection

Technological advancements over the years have increased the use of Radar technology in the field of robotics, especially in marine robotics to aid obstacle detection algorithms. Obstacle detection comprises of an analytical process in which different algorithms are applied to the field of study to determine the range of objects that are within the reach of a robot. Radar signal analysis and target detection in conjunction with target tracking are attributes required for autonomous marine navigation. The paper presents a model which converts optimal estimates of radar range values for each range spectra into multiple targets down-range and also presents an approach for power range spectra (range bins) prediction using the radar range equation with adequate information of the signal-to-noise ratio (SNR) of the radar. Obstacle detection in the presence of noise raises certain probabilities of false alarms. Target characteristics are simulated and these are fluctuating targets and non fluctuating targets. Analytical models, simulations and techniques of obstacle detection for autonomous marine craft navigation using a continuous wave radar system were points of discussion in the paper.