Key Application Scenarios: Pose Measurement and Positioning Guidance

(Applicable to: Vessel Pose, Vehicle Positioning Guidance and Navigation)

Application Requirements:

Vessel Pose Measurement: During automated loading and unloading operations, it is necessary to accurately measure the position of the vessel after berthing, as well as pose changes (roll, pitch, yaw, heave) caused by tides and loading/unloading. This enables automatic adjustment of the working position of loading arms or grab buckets to ensure operational accuracy and safety.

Vehicle Positioning Guidance and Navigation: Enables AGVs, autonomous trucks, transfer vehicles, etc., to achieve autonomous and precise navigation and positioning in complex environments such as ports, terminals, storage yards, and mining areas to complete material transportation tasks.

Typical Deployment and Architecture:

Vessel Pose: Sensors are typically installed on the quay or loading/unloading equipment (such as ship loaders) to scan the hull contour or specific reflective targets, and calculate the six-degree-of-freedom pose of the vessel through geometric analysis.

Vehicle Navigation: Sensors (such as Ouster OS0/OS1/OSDome) are installed on the top or around the vehicle to build real-time environmental maps and perform simultaneous localization (SLAM), or position based on pre-built high-precision maps. The system usually fuses multi-source information from IMU, GPS (if available), wheel speed meters, etc., to improve positioning accuracy and robustness.

Traditional Solutions and Pain Points:

GPS/GNSS: In environments with large metal structures (cranes, factories) or topographic obstructions such as ports and mines, satellite signals are easily interfered with or lost, resulting in reduced positioning accuracy (meter-level errors) or interruptions, failing to meet the centimeter-level accuracy requirements of automated operations. It cannot be used in indoor or underground environments.

Infrastructure-Based Positioning Systems: Such as embedding magnetic nails, RFID tags in the ground or laying guide wires, etc., which require large-scale infrastructure investment and continuous maintenance, have poor deployment flexibility, and are difficult to adapt to changing operational environments.

2D LiDAR SLAM: In feature-sparse, repetitive-scene, or open environments such as bulk material storage yards, positioning drift or failure is prone to occur. It is sensitive to dynamic obstacles and cannot perceive 3D structure information, limiting its application in complex environments.

Visual SLAM (VSLAM): Extremely sensitive to environmental factors such as light changes, weather (rain, fog, snow), and dust. It is prone to failure in areas with little or repetitive texture (such as large-area material piles, walls).

Pure Inertial Navigation System (INS): Without external information (such as GPS, LiDAR SLAM) for correction, its position and pose errors will accumulate and increase rapidly over time.

Performance Requirements:

For vehicle navigation, centimeter-level positioning accuracy and high update rate are required, capable of stable and reliable operation in dynamic and complex environments, and able to build or update maps in real time.

For vessel pose measurement, high-precision 3D scanning capability is required.

Ouster 3D Digital LiDAR Alternative Benefits:

Robust 3D SLAM/Positioning: The high-density and high-precision 3D point clouds generated by Ouster LiDAR provide rich geometric features, enabling 3D LiDAR-based SLAM algorithms (such as LIO-SAM, LVI-SAM) to achieve robust positioning and mapping in complex, dynamic, and feature-sparse environments where 2D LiDAR or visual SLAM struggle. Ouster sensors have been used in the reference implementations of LIO-SAM/LVI-SAM.

Tightly Coupled LIO/LVI Advantages: The IMU integrated in Ouster LiDAR and strictly time-space synchronized with point cloud data is the key to high-performance LIO (LiDAR-Inertial Odometry)/LVI (LiDAR-Visual-Inertial Odometry) SLAM. Tightly coupled fusion algorithms can make full use of the high-frequency motion information provided by the IMU to assist LiDAR pose estimation, significantly improving accuracy and robustness, and effectively compensating for motion distortion. This is more effective than fusing separate LiDAR and IMU data later.

Multimodal SLAM (LVI-SAM): The camera-like image data (ambient light/intensity map) simultaneously output by Ouster LiDAR enables LVI-SAM. By fusing geometric and visual features, it is expected to achieve higher robustness than pure LIO in certain specific scenarios (such as environments with degraded geometric structures but rich visual textures).

All-Weather Environmental Adaptability: Strong adaptability to dust, weather, and light changes ensures continuous and reliable operation of the navigation system under various harsh working conditions.

Optimized Sensor Form Factor: The wide vertical field of view of OS0/OSDome is very suitable for installation on vehicle platforms, enabling simultaneous perception of the ground and upper environments. The zero-blind-zone feature of REV7 enhances close-proximity obstacle detection capability, which is crucial for safe navigation.

Precise Vessel Pose Measurement: High-resolution point clouds can accurately capture hull feature points or installed cooperative targets, thereby accurately calculating the six-degree-of-freedom pose of the vessel. The MRA/QCA system and Avikus case both demonstrate the potential of Ouster LiDAR in vessel-related applications.

Integrated Case Analysis:

Autonomous Navigation of AGVs in Bulk Material Storage Yards: Achieving autonomous navigation of AGVs in unstructured bulk material storage yards is more challenging than in warehouse environments, with unstable GPS signals and prone failure of 2D SLAM. It is recommended to use Ouster OS0, OS1, or OSDome REV7 sensors combined with LIO-SAM or LVI-SAM algorithms. Their powerful 3D SLAM capability, adaptability to harsh environments, and integrated IMU enable high-precision and high-robustness autonomous navigation. TAGE's successful use of Ouster OS1 for autonomous mining trucks in mining areas proves its feasibility in similar scenarios. AGV solution providers such as BlueBotics also use Ouster LiDAR.

Dynamic Vessel Pose Measurement: Automated loading and unloading systems need to accurately grasp vessel poses in real time to compensate for their movements. Manual alignment or simple sensors cannot meet the requirements. It is recommended to install Ouster OS1 or OS2 REV7 sensors on quay cranes or loading arms to continuously scan the hull. High-resolution 3D point clouds can accurately measure the position and pose changes of the hull. Their high reliability and robustness in port environments are key advantages. Avikus' use of Ouster sensors for vessel navigation and obstacle avoidance, as well as the MRA/QCA system, indirectly supports the applicability of Ouster LiDAR in this application.