Static Object Scoring in Maps for Autonomous Vehicle Localization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in accurately determining their location within environments due to the lack of effective methods to differentiate between static and dynamic objects, which affects their navigation and route planning.
Innovation Solution
The system assigns static scores to elements in the environment based on their likelihood of immobility, using these scores to inform localization and route planning, with sensors like LIDAR and cameras generating maps that include static scores for boundaries, landmarks, and objects, and a processing device adjusting scores over time based on conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the autonomous vehicle uses all detected objects for localization, then the quantity of localization data increases, but the accuracy of location determination decreases due to inclusion of dynamic objects
Solution Approach 1:
The patent segments the environment into static and dynamic objects by assigning mobility scores to each detected object. This segmentation allows the system to selectively use only high-score static objects for localization, filtering out dynamic objects that would reduce accuracy. The segmentation is achieved through continuous monitoring of object positions and classification based on mobility thresholds.
Solution Approach 2:
The patent applies local quality by differentiating the treatment of different objects based on their individual mobility characteristics. Static objects (high mobility scores) are selected for localization while dynamic objects (low mobility scores) are excluded. This selective application of localization data based on local object properties improves overall location determination accuracy without requiring all detected objects.
2Reliability
If the system continuously updates object mobility scores, then the reliability of localization increases, but the energy consumption increases
Solution Approach 1:
The patent implements periodic action by updating mobility scores at specific intervals rather than continuously. The system monitors object positions periodically and updates scores only when changes exceed certain thresholds or at predetermined time intervals. This periodic updating maintains localization reliability by detecting significant environmental changes while reducing energy consumption compared to continuous monitoring.
Solution Approach 2:
The system applies self-service by using the autonomous vehicle's own sensor data and processing capabilities to maintain mobility scores without requiring external intervention or additional dedicated hardware. The vehicle leverages its existing LIDAR, cameras, and processors to perform self-localization using static objects, eliminating the need for separate localization systems and reducing overall energy requirements.
3Adaptability or versatility
If the autonomous vehicle uses low mobility score objects for localization, then the adaptability to dynamic environments improves, but the measurement precision of location decreases
Solution Approach 1:
The patent applies dynamics by making the set of localization objects dynamic rather than static. The system continuously evaluates mobility scores and adjusts which objects are used for localization based on current environmental conditions. When dynamic objects become static (e.g., a moving forklift parks), they are automatically added to the localization set, providing adaptability while maintaining precision through score-based selection.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An example system includes a sensor for obtaining information about an object in an environment and one or more processing devices configured to use the information in generating or updating a map of the environment. The map includes the object and boundaries or landmarks in the environment. The map includes a static score associated with the object. The static score represents a likelihood that the object will remain immobile within the environment. The likelihood may be between certain immobility and certain mobility.