Source-Agnostic Vehicle Trajectories Without Coordinate Translation Errors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing approaches for collecting prior agent trajectories face challenges due to the high cost and limited availability of LiDAR-based sensor systems, leading to inaccurate and inefficient collection methods, particularly when translating trajectories from source-specific to destination coordinate frames, which introduces translation errors and inefficiencies.
Innovation Solution
A framework for deriving and storing agent trajectories in a source-agnostic coordinate frame, such as the Earth-Centered Earth-Fixed (ECEF) frame, which eliminates the need for two-step translations, maintains accuracy, and allows for efficient storage and use across different sensor systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If LiDAR-based sensor systems are used to collect agent trajectories, then measurement precision is improved, but device complexity and cost increase
Solution Approach 1:
The patent introduces a source-agnostic coordinate frame as an intermediary representation that mediates between different sensor systems. Instead of requiring all sensors to directly output in a unified frame, the system translates trajectories from their source coordinate frames into this intermediate source-agnostic frame, eliminating the need for complex direct coordination between multiple sensor systems while maintaining measurement precision.
Solution Approach 2:
The source-agnostic coordinate frame serves as a universal representation that can accommodate trajectories from multiple different sensor systems (LiDAR, camera-based, radar-based, etc.). This universal frame allows the same trajectory processing pipeline to handle data from diverse sensor sources without requiring sensor-specific handling, reducing overall system complexity while maintaining high measurement precision through proper coordinate transformations.
2Adaptability or versatility
If trajectories are translated between different coordinate frames, then adaptability is improved, but measurement precision deteriorates due to translation errors
Solution Approach 1:
The source-agnostic coordinate frame acts as an intermediary that receives trajectories from various source coordinate frames and translates them into a unified representation. This intermediate step consolidates all coordinate frame transformations into a single managed process, reducing cumulative translation errors that would occur if trajectories were directly translated between multiple different coordinate frames without this intermediate reference.
Solution Approach 2:
The system performs preliminary translation of trajectories from their source coordinate frames into the source-agnostic coordinate frame before any further processing or use. This preliminary action establishes a common reference state for all trajectories, ensuring that subsequent operations work with consistent coordinate representations and preventing accuracy degradation from repeated transformations.
3Device complexity
If multiple sensor systems with different accuracy levels are used, then device complexity is reduced, but measurement precision varies across data sources
Solution Approach 1:
The patent changes the parameter of coordinate frame representation to source-agnostic, which allows the system to handle trajectories from sensor systems with varying accuracy levels uniformly. By translating all trajectories into this common reference frame and storing them with associated accuracy metadata, the system maintains measurement precision consistency across diverse sensor sources while avoiding the complexity of sensor-specific processing pipelines.
4Ease of operation
If source-specific coordinate frames are used for each sensor system, then ease of operation is improved for each individual sensor, but adaptability across different sensor systems deteriorates
Solution Approach 1:
The source-agnostic coordinate frame serves as an intermediary layer that preserves the simplicity of sensor-specific processing while enabling cross-sensor compatibility. Each sensor system can continue to operate independently in its own coordinate frame for ease of processing, but all trajectories are translated to the source-agnostic frame for unified handling, thus maintaining both operational simplicity and adaptability across different sensor systems.
Data Source
AI summary
Examples disclosed herein involve a computing system configured to (i) obtain (a) a first set of sensor data captured by a first sensor system of a first vehicle that indicates the first vehicle's movement and location with a first degree of accuracy and (b) a second set of sensor data captured by a second sensor system of a second vehicle that indicates the second vehicle's movement and location with a second degree of accuracy that differs from the first degree of accuracy, (ii) based on the first set of sensor data, derive a first trajectory for the first vehicle that is defined in terms of a source-agnostic coordinate frame, (iii) based on the second set of sensor data, derive a second trajectory for the second vehicle that is defined in terms of the source-agnostic coordinate frame, and (iv) store the first and second trajectories in a database of source-agnostic trajectories.