Subsea Vehicle Pilot Assist With AI Object Tracking and Autonomy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Subsea operations face challenges with limited communication bandwidth and restricted movement due to tethered connections, and untethered autonomous underwater vehicles experience lower communication rates and intermittent interruptions, compounded by environmental factors like noise and limited sensor visibility.
Innovation Solution
A system and method utilizing machine learning algorithms and multi-modal approaches for subsea ROV/AUV operations, incorporating hardware and software to assist operators with object tracking, control, and anomaly detection, enabling interactive and autonomous navigation with features like augmented views, smart filters, and continuous anomaly detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If the vehicle is physically tethered via cable to surface equipment, then high bitrate connectivity is achieved, but the movement of the vehicle is restricted
Solution Approach 1:
The system divides control functions into two segments: high-level autonomous navigation handled by the AUV's onboard processor, and low-level teleoperated control for精细 manipulation. This segmentation allows the vehicle to operate autonomously without tether constraints while maintaining high-bandwidth communication only when needed for critical operations.
Solution Approach 2:
The AUV performs self-navigation and self-positioning using onboard sensors (inertial navigation, depth sensors, compass) and autonomous obstacle detection algorithms. This self-service capability eliminates the need for continuous tethered guidance, freeing the vehicle from movement restrictions while maintaining operational effectiveness.
2Adaptability or versatility
If untethered autonomous underwater vehicles are used, then vehicle movement freedom is improved, but communication rate is reduced and may be interrupted intermittently
Solution Approach 1:
The AUV executes preliminary autonomous actions (navigation to target, obstacle avoidance, positioning) before requiring communication. The vehicle independently completes navigation tasks using pre-programmed algorithms and onboard sensors, only establishing communication when human oversight or intervention is needed, thus minimizing communication interruptions.
Solution Approach 2:
The system implements autonomous feedback loops using onboard sensors to continuously monitor position, orientation, and environmental conditions. The vehicle adjusts its course and maneuvers based on real-time sensor feedback, maintaining autonomous operation without requiring continuous communication feedback from the surface.
3Ease of operation
If traditional sensor-based navigation is used, then basic navigation capability is achieved, but performance deteriorates in noisy and cluttered environments with limited visibility
Solution Approach 1:
The system merges multiple independent navigation systems: inertial navigation (accelerometers, gyroscopes), acoustic positioning (USBL/LBL systems), visual odometry (stereo cameras), and sonar-based obstacle detection. This multi-sensor fusion approach provides redundant navigation capabilities that remain reliable even when individual sensors fail or environmental conditions degrade.
Solution Approach 2:
The navigation system uses composite sensing approaches, combining data from heterogeneous sensor types (inertial, acoustic, optical, sonar) with different operating principles. This composite sensing strategy ensures navigation reliability across diverse environmental conditions by leveraging the complementary strengths of each sensor modality.
Data Source
AI summary
A method for controlling a subsea vehicle. The method includes receiving sensor data representing a subsea environment from one or more sensors of the subsea vehicle. The method identifies one or more objects present in the subsea environment based on the sensor data using an artificial intelligence machine. The method transmits at least a portion of the sensor data, including an identification of the one or more objects, to a user interface. The method includes receiving a requested vehicle task from the user interface. The requested vehicle task being selected by a user via the user interface. The method performs the requested vehicle task without vehicle position control from the user.