LiDAR Target Tracking with Iterative Reference Model Adaptation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing LIDAR sensors face inaccuracies in detecting and tracking non-cooperative spacecraft at long distances due to noise behavior and lack of distinctive recognition elements, leading to significant position estimation errors, especially beyond 50 meters.
Innovation Solution
A method involving continuous emission of laser pulses by a LIDAR sensor to generate high-resolution 3D point clouds, using a noisy reference model adapted iteratively from initializing scans, and applying ICP algorithms for position estimation, allowing for accurate tracking beyond 50 meters by compensating for sensor noise and improving the reference model continuously.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Length of stationary object
If LiDAR sensors are used for long-range detection of non-cooperative spacecraft, then detection range is extended, but measurement precision deteriorates due to noise characteristics
Solution Approach 1:
The patent applies preliminary action by performing an initial high-resolution acquisition scan to create a detailed reference model of the target object before conducting tracking scans. This reference model is created in advance and used to guide subsequent tracking operations, enabling accurate position estimation at long ranges by providing a noise-resistant reference for comparison with noisy tracking data.
Solution Approach 2:
The patent segments the detection process into two distinct phases: acquisition phase (creating high-resolution reference model) and tracking phase (comparing noisy tracking scans against the reference model). This segmentation allows each phase to be optimized independently - the acquisition phase captures detailed geometry while the tracking phase focuses on efficient position estimation using the pre-built reference.
2Device complexity
If average value calculation is used to determine relative position from point clouds, then calculation simplicity is maintained, but reliability deteriorates due to highly unevenly distributed points
Solution Approach 1:
The patent implements feedback by iteratively comparing tracking scan points with the reference model and using the results to refine position estimates. The ICP algorithm provides feedback loops where the transformation parameters are continuously adjusted based on the comparison between tracked points and reference model points, improving reliability without requiring complex calculations.
Solution Approach 2:
The patent uses copying by creating a reference model that replicates the geometric structure of the target object from initial high-resolution scans. This reference copy is then used repeatedly for comparison with tracking scans, allowing reliable position estimation through simple point-to-point or point-to-surface distance calculations rather than complex averaging of unevenly distributed points.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach significantly reduces position estimation fluctuations, achieving accuracy of less than 1 meter at 1.5 kilometers with noisy measurements, enhancing the reliability of LIDAR sensors for autonomous tracking in space missions.
Implementation Method 1
The LIDAR sensor continuously emits laser pulses at the target object over a predetermined area and generates a 3D point cloud from measurement points reflected by the target object
Implementation Method 2
a 3D point cloud is generated at predetermined time intervals from measurement points reflected by the target object
Data Source
Figure 1
AI summary
The invention relates to a method for detecting and autonomously tracking a target object, in particular a non-cooperative spacecraft, using a LIDAR sensor, wherein the LIDAR sensor continuously emits laser pulses at the target object over a predetermined area and generates a 3D point cloud from measurement points reflected by the target object at predetermined time intervals, and a current relative position of the target object relative to the LIDAR sensor is calculated at predetermined time intervals using position-estimating algorithms for six degrees of freedom of the target object.To extend the detection range of target objects to greater distances, a high-resolution initial point cloud is acquired in a first step over several predefined time intervals. In a second step, a reference model with a small number of measurement points is generated from the initial point cloud. In subsequent iterative steps, position clouds are determined at the given time intervals and compared with the reference model using algorithms. Position information of the target object is calculated, with the reference model being continuously adapted to changes in the point clouds.