Autonomy Map Constraint Layering for Fleet Operators
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems for autonomous vehicles (AVs) require labor-intensive processes to update autonomy maps, making it inefficient to incorporate granular information and changes in route environments, such as road construction, into existing navigation policies.
Innovation Solution
A unified document model enables fleet operators to generate and update map constraints on existing autonomy maps without creating new ones, using a map constraint interface that categorizes constraints by sequenced operations, allowing for flexible configuration and deployment of navigational, perception, and prediction constraints.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If traditional autonomy maps are updated by re-recording and relabeling, then map accuracy can be maintained, but the process becomes highly labor-intensive and inefficient
Solution Approach 1:
The patent segments the autonomy map update process by introducing constraint layers that can be independently created, modified, and applied to specific route portions without re-recording entire maps. Constraints are divided into navigational constraints (affecting motion planning) and perception constraints (affecting object detection), allowing granular updates to specific map regions while preserving the rest of the map structure.
Solution Approach 2:
The system performs preliminary actions by pre-defining constraint templates and layers that can be rapidly deployed to autonomy maps. Fleet operators can configure constraint layers in advance and apply them to multiple route portions simultaneously, eliminating the need for time-consuming re-recording and relabeling operations for each update scenario.
2Reliability
If granular information about route environment changes is incorporated into autonomy maps, then navigation safety is improved, but the complexity of map management increases
Solution Approach 1:
The patent extracts environmental constraints from the main autonomy map structure and places them into separate, manageable constraint layers. This separation allows granular information about route changes (construction zones, temporary restrictions, hazard locations) to be incorporated without complicating the core map data structure. Constraint layers can be independently configured, validated, and applied.
Solution Approach 2:
The system adds a new dimension to map management by introducing a temporal and hierarchical layering structure. Constraint layers operate as an additional dimension overlaying the base autonomy map, allowing multiple constraint sets to coexist at different route portions. This dimensional approach enables sophisticated navigation safety rules without increasing fundamental map management complexity.
3Measurement precision
If new autonomy maps are created to reflect environment changes, then navigation accuracy is maintained, but the time and resources required for map creation increase
Solution Approach 1:
The patent uses copying by allowing constraint layers to be replicated and applied across multiple route portions. Instead of creating entirely new autonomy maps for each environmental change scenario, the system copies and adapts existing constraint templates to new locations. This maintains navigation accuracy through consistent constraint application while dramatically reducing map creation time.
Solution Approach 2:
The system maintains navigation accuracy by changing parameters within constraint layers (such as constraint type, affected route portions, temporal validity) rather than creating new maps. Fleet operators can modify constraint parameters to reflect environmental changes, and these parameter adjustments are automatically integrated into the existing autonomy map structure, preserving accuracy without time-consuming map recreation.
Data Source
AI summary
A computing system can generate a map constraint interface enabling a fleet operator to update map constraints for autonomous vehicles (AVs). The map constraint interface can comprise a unified document model enabling the fleet operator to configure a set of constraint layers of autonomy maps utilized by the AVs. Each constraint layer can include a toggle feature that enables the fleet operator to enable and disable the constraint layer. The system can receive, via the map constraint interface, a set of inputs configuring the set of constraint layers of the one or more autonomy maps, compile a set of updated map constraints, corresponding to the configured set of constraint layers, into a document container, and output the document container to a subset of the AVs to enable the subset of AVs to integrate the set of updated map constraints with the autonomy maps.