Sensor autonomy

To the end user, an AUV is nothing more than a platform carrying the sensors (or other payloads) that perform the actual objective of the mission; data collection. In fully autonomous HUGIN missions, the performance of the onboard sensors and other subsystems are normally checked during the initial phase of the mission, where the AUV is within communication range of the surface vessel. Using either acoustic or RF links, the operator can ensure that all subsystems perform as assumed during the planning phase -and take corrective action as needed. If conditions are known to vary, different sensor settings may be pre-programmed for different parts of the mission plan.

However, when operating autonomously in areas where the environmental conditions are unknown or rapidly varying, this may not be sufficient and the AUV will need to re-adjust the sensor configuration autonomously.

Some sensors include some functionality for automatic adaptation to varying conditions; some are very simple and have no parameters that can be changed. However, most sensors available today are designed based on the assumption that a human operator will monitor the sensor data and adjust the sensor parameters as needed. This is in particular true for more advanced sensors such as side scan sonars, multibeam echo sounders, synthetic aperture sonars and optical imaging systems.

Over the following paragraphs, we will use the Kongsberg HISAS 1030 interferometric synthetic aperture sonar (SAS) as an example (Fossum et al., 2008). This sensor is used for very high resolution imaging (2-5 cm both along- and across-track) and high resolution bathymetry out to approximately 10 times AUV altitude.

HISAS 1030 transmits wide-beam, wide-band pulses to port and starboard side at regular intervals; typically, a few Hz. Signals reflected back from the seafloor and from objects in the water volume are received in two long multi-element receiver arrays, as well as in the transmitter. A complex signal processing chain, working on data from a number of consecutive pings, transforms the raw data into sonar imagery and bathymetry (Hagen et al., 2001)(Hansen et al., 2003).

The performance of a SAS is limited by a wide range of factors - AUV altitude, stability of motion, navigation accuracy, self noise, and sound speed variations, to name a few. In shallow water, performance is often limited by acoustic multipath; i.e., simultaneous reception of signals that have travelled different paths from transmitter to receiver (reflected from the seafloor, from the surface, seafloor then surface, surface then seafloor, etc). A fully autonomous SAS would analyse the returned signal and, based on the sonar data and knowledge of the environment, tune operation to optimize performance. The most obvious candidate parameters to modify include the transmission beam width and direction, and the AUV altitude. After applying these changes, the autonomy system may determine that performance is still inadequate; e.g., the actual range to which good sonar imagery can be produced is less than required to achieve full bottom coverage. This will then trigger replanning of the mission, by spacing survey lines tighter.

Learn Photoshop Now

Learn Photoshop Now

This first volume will guide you through the basics of Photoshop. Well start at the beginning and slowly be working our way through to the more advanced stuff but dont worry its all aimed at the total newbie.

Get My Free Ebook

Post a comment