UMV Obstacle Avoidance Using Camera Classification and Sensor Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current unmanned maritime vehicles (UMVs) lack onboard systems to detect obstacles both above and below water, relying solely on cameras and external human input for navigation, which limits their autonomous operation and ability to assess their surroundings effectively.
Innovation Solution
A system equipped with image sensors, storage elements, and processing units that identify obstacles through video images, determine their distance based on object classes, and adjust navigational controls to avoid them, incorporating multiple sensor inputs like SONAR and LiDAR for robust environmental assessment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If UMVs rely solely on cameras and external human input for navigation, then device complexity is reduced, but autonomous operation capability and obstacle detection accuracy deteriorate
Solution Approach 1:
The system divides the sensing and processing functions into separate modules: image sensors for visual detection, SONAR for underwater obstacle detection, LiDAR for precise ranging, and a processing unit that integrates data from all sensors. This segmentation allows each component to specialize in specific detection tasks, improving autonomous operation while maintaining manageable system complexity through modular design.
Solution Approach 2:
The processing unit serves multiple functions: it processes data from image sensors, SONAR, and LiDAR; performs obstacle classification; calculates distance and bearing; and generates navigation commands. This multi-functionality consolidates what could be separate complex systems into a single integrated unit, enhancing autonomous capability without proportionally increasing overall system complexity.
2Measurement precision
If UMVs use only cameras for obstacle detection, then device complexity is minimized, but measurement precision and detection reliability deteriorate
Solution Approach 1:
The system merges multiple sensing technologies (image sensors, SONAR, LiDAR) into an integrated obstacle detection system. Each sensor type compensates for the limitations of others: image sensors provide visual identification, SONAR detects underwater obstacles invisible to cameras, and LiDAR provides precise distance measurements. This combination significantly improves measurement precision while the shared processing unit manages the complexity of integrating multiple sensor types.
Solution Approach 2:
The processing unit acts as an intermediary that receives raw data from multiple sensor types, integrates and correlates this information, and produces unified obstacle detection results. This intermediary function allows the system to leverage the strengths of each sensor type while managing the complexity of multi-sensor integration through centralized data processing and fusion algorithms.
3Productivity
If UMVs depend on third-party inputs like AIS and human interaction, then device complexity is reduced, but productivity and operational independence deteriorate
Solution Approach 1:
The UMV performs self-service through autonomous obstacle detection and avoidance. The integrated sensor system and processing unit enable the vehicle to independently identify obstacles, determine their positions and distances, classify them as hazards, and generate appropriate navigation commands without requiring continuous human input or reliance on external AIS systems. This self-service capability significantly improves operational efficiency and productivity by enabling autonomous operation.
Solution Approach 2:
The system implements continuous feedback loops where sensor data is constantly monitored, processed, and used to adjust navigation in real-time. The processing unit receives ongoing input from image sensors, SONAR, and LiDAR, dynamically updates obstacle positions and threats, and continuously generates updated navigation commands. This real-time feedback mechanism enables responsive autonomous operation, improving productivity by eliminating delays associated with external communication and human decision-making.
Data Source
AI summary
A method and system for detecting and avoiding an obstacle of an autonomous unmanned maritime vehicle (UMV) traveling in an initial direction are described. The method and system include receiving a video image from at least one image sensor, the video image containing an object that has been identified as an obstacle, determining if the object can be associated with an object class from the plurality of predetermined object classes, accessing a mean height value for the object class if it determined that the object can be associated with an object class from the plurality of predetermined object classes, determining a distance between the UMV and the object based on a height of the object as displayed in the video image and the mean height for the object class, and automatically adjusting the navigational control of the autonomous UMV to travel in an adjusted direction.