System and method for autonomous exploration for mapping underwater environments
Embodiments of the present disclosure are directed towards a system and method for performing an inspection of an underwater environment. Embodiments may include providing an autonomous underwater vehicle (“AUV”) and performing an inspection of an underwater environment using the AUV. Embodiments may further include acquiring real-time sensor data during the inspection of the underwater environment and applying an active simultaneous localization and mapping (“SLAM”) algorithm during the inspection, wherein applying includes estimating one or more virtual landmarks based upon, at least in part, at least one past measurement and a current estimate of AUV activity.
1. A method for performing an inspection of an underwater environment comprising:
executing an autonomous exploration, in whole or in part, by an autonomous underwater vehicle (“AUV”);
performing an inspection of an underwater environment using the AUV;
constructing, by a processor, one or more candidate paths;
acquiring, by the processor, real-time sensor data during the inspection of the underwater environment;
applying, by the processor, an active simultaneous localization and mapping (“SLAM”) algorithm during the inspection,
wherein applying includes estimating one or more virtual landmarks based upon, at least in part, at least one past measurement and a current estimate of AUV activity, and
wherein applying includes applying an Expectation-Maximization exploration algorithm with a utility function explicitly addressing a map accuracy by identifying and periodically re-observing features in the underwater environment for navigating a previously identified virtual landmark of the estimated one or more virtual landmarks;
evaluating, by the processor, a localization error of the one or more virtual landmarks resulting from the one or more candidate paths; and
determining, by the processor, a best path based on the localization error and a map information gain represented by the utility function.
2. The method for performing an inspection of an underwater environment of claim 1 , wherein the AUV includes a sensor configuration consisting of one or more multibeam sonars, lidars, and cameras.
3. The method for performing an inspection of an underwater environment of claim 1 , further comprising:
using one or more fiducial markers to facilitate localization and mapping of an infrastructure by the AUV.
4. The method for performing an inspection of an underwater environment of claim 1 , further comprising:
segmenting three-dimensional (“3D”) data associated with the real-time sensor data for at least one of segment, object, and place recognition.
5. The method for performing an inspection of an underwater environment of claim 1 , further comprising:
executing the best path by the AUV.
6. The method for performing an inspection of an underwater environment of claim 5 , further comprising:
determining termination criteria.
7. The method for performing an inspection of an underwater environment of claim 6 , further comprising:
using one or more fiducial markers to facilitate localization and mapping of an infrastructure by the AUV;
if the termination criteria are met, completing the mapping.
8. A system for performing an inspection of an underwater environment comprising:
an autonomous underwater vehicle (“AUV”) configured to execute an autonomous exploration, in whole or in part, and to perform an inspection of an underwater environment;
one or more sensors associated with the AUV configured to acquire real-time sensor data during the inspection of the underwater environment; and
at least one processor associated with the AUV configured to:
construct one or more candidate paths;
apply an active simultaneous localization and mapping (“SLAM”) algorithm during the inspection, wherein applying includes estimating one or more virtual landmarks based upon, at least in part, at least one past measurement and a current estimate of AUV activity, and wherein applying includes applying an Expectation-Maximization exploration algorithm with a utility function explicitly addressing a map accuracy by identifying and periodically re-observing features in the underwater environment for navigating a previously identified virtual landmark of the estimated one or more virtual landmarks;
evaluate a localization error of the one or more virtual landmarks resulting from the one or more candidate paths; and
determine a best path based on the localization error and a map information gain represented by the utility function.
9. The system for performing an inspection of an underwater environment of claim 8 , wherein the one or more sensors includes a sensor configuration consisting of one or more multibeam sonars, lidars, and cameras.
10. The system for performing an inspection of an underwater environment of claim 8 , wherein the at least one processor is further configured to use one or more fiducial markers to facilitate localization and mapping of an infrastructure by the AUV.
11. The system for performing an inspection of an underwater environment of claim 8 , wherein the at least one processor is further configured to segment three-dimensional (“3D”) data associated with the real-time sensor data for at least one of segment, object, and place recognition.
12. The system for performing an inspection of an underwater environment of claim 8 , wherein the at least one processor is further configured to execute the best path by the AUV.
13. The system for performing an inspection of an underwater environment of claim 12 , wherein the at least one processor is further configured to determine termination criteria.
14. The system for performing an inspection of an underwater environment of claim 13 , wherein the at least one processor is further configured to use one or more fiducial markers to facilitate localization and mapping of an infrastructure by the AUV, and wherein if the termination criteria are met, the at least one processor is further configured to complete the mapping.