Pose Estimation Using Long Range Retroreflective Features
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in achieving high accuracy for geographic location and heading estimation due to noise and uncertainty in inertial pose estimation and localization, which are often limited to near-range information.
Innovation Solution
The method involves using a first sensor to generate a map of the vehicle's environment within its range and a second sensor with a greater range to identify reference objects with specific retroreflective, brightness, or intensity characteristics, correlating these objects to a stored map to refine the vehicle's geographic location and heading.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If inertial pose estimation systems and map-based localization approaches are used, then the vehicle can estimate its geographic location and heading, but noise and uncertainty in both stages make it difficult to achieve very high accuracy results
Solution Approach 1:
The patent introduces an intermediary process that combines inertial pose estimation with map-based localization through a unified optimization framework. This intermediary approach reconciles the two estimation methods by jointly optimizing their combination, thereby reducing the noise and uncertainty that arise when they are used separately. The system uses this intermediary optimization to produce more reliable and accurate location and heading estimates.
2Measurement precision
If near-range information is focused on to build local maps, then localization can be performed, but the sensor cannot detect objects with sufficient density at long range
Solution Approach 1:
The patent merges near-range sensor data with long-range sensor data into a unified localization framework. By combining these data sources, the system overcomes the limitation of individual sensors that cannot detect objects with sufficient density at long ranges. The merged information from both near and far sensors enables accurate localization while extending the effective detection range beyond what a single sensor could achieve.
3Device complexity
If a single sensor is used for environmental mapping, then the system is simpler, but the sensor range limits the ability to identify reference objects for refining location estimates
Solution Approach 1:
The patent implements a dynamic sensor system that adapts its configuration based on operational requirements. The system can dynamically switch between using a single sensor for simpler operations and utilizing multiple sensors with different ranges when high-precision location estimation is needed. This dynamic approach allows the system to maintain simplicity when possible while achieving high accuracy when required by adjusting the sensor configuration in real-time.
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 enhances the accuracy of location and heading estimation by utilizing long-range data to correct initial estimates, reducing errors and improving navigation precision.
Implementation Method 1
reference objects having particular retroreflective, brightness, or intensity characteristics
Data Source
AI summary
Aspects of the present disclosure relate to using an object detected at long range to increase the accuracy of a location and heading estimate based on near range information. For example, an autonomous vehicle may use data points collected from a sensor such as a laser to generate an environmental map of environmental features. The environmental map is then compared to pre-stored map data to determine the vehicle's geographic location and heading. A second sensor, such as a laser or camera, having a longer range than the first sensor may detect an object outside of the range and field of view of the first sensor. For example, the object may have retroreflective properties which make it identifiable in a camera image or from laser data points. The location of the object is then compared to the pre-stored map data and used to refine the vehicle's estimated location and heading.