Automated Driving Path Selection Using Roadway Cue Grouping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle navigation systems face difficulties in accurately identifying and following roadway paths due to limited sensor perspectives, occluded or worn lane markers, and dynamic environmental conditions, leading to inadequate contextual awareness and navigation challenges.
Innovation Solution
A tracking system that acquires sensor data to identify roadway elements, characterizes their characteristics, groups them using clustering algorithms, and applies a confidence heuristic to determine priority metrics, enabling improved path planning and navigation by utilizing environmental cues.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional sensors are used to detect roadway elements, then the system can operate with simple hardware, but the measurement precision and reliability deteriorate when lane markers are occluded or worn
Solution Approach 1:
The patent combines multiple roadway element detections (lane markers, surrounding vehicles, curbs, signs) into a unified path identification system. By merging these different cues and analyzing them collectively through clustering algorithms, the system achieves higher measurement precision in identifying navigable paths, especially when some elements are occluded or worn.
Solution Approach 2:
The system transitions from detecting single-dimensional lane markers to multi-dimensional environmental cues by incorporating vehicle trajectories, curb positions, and sign locations. This dimensional expansion allows the system to infer path information from multiple sources, improving accuracy when traditional lane markers are insufficient.
2Loss of information
If sensors acquire data from a limited perspective at road-level, then the device complexity remains low, but the loss of information increases due to occlusions and weather conditions
Solution Approach 1:
The sensor system is designed to detect multiple types of roadway elements (lane markers, vehicles, curbs, signs) simultaneously with a single observation. This multi-functional approach reduces information loss by capturing diverse contextual cues from the same data acquisition event, enabling comprehensive environmental awareness without requiring multiple specialized sensors.
Solution Approach 2:
The system performs preliminary clustering and grouping of roadway elements based on their spatial and temporal characteristics. By pre-processing the sensor data to identify patterns and relationships among different elements before path determination, the system preserves contextual information that would otherwise be lost in raw sensor data.
3Measurement precision
If the system processes sensor data in real-time with constrained computational time, then the productivity is high, but the measurement precision deteriorates due to insufficient analysis time
Solution Approach 1:
The path determination process is segmented into distinct stages: roadway element detection, characteristic extraction, clustering/grouping, and priority metric calculation. This segmentation allows the system to process data in manageable chunks with optimized computational requirements at each stage, maintaining both precision and processing speed.
Solution Approach 2:
The system applies partial processing by focusing computational resources on the most relevant roadway elements and characteristics for path determination. Rather than analyzing all possible features in equal detail, the system prioritizes key elements (such as clustered vehicle trajectories and prominent lane markers) to achieve sufficient measurement precision within constrained time.
Data Source
AI summary
Systems, methods, and other embodiments described herein relate to improving identification of a path for an ego vehicle on a roadway. In one embodiment, a method includes, in response to acquiring sensor data from at least one sensor of the ego vehicle about a surrounding environment, identifying roadway elements from the sensor data as cues about the path. The roadway elements include one or more of lane markers of the roadway and surrounding vehicles. The method includes grouping the roadway elements into two or more groups according to characteristics of roadway elements indicating common curvatures. The method includes analyzing the two or more groups according to a confidence heuristic to determine a priority group from the two or more groups that corresponds with a trajectory of the ego vehicle. The method includes providing an identifier for the priority group to facilitate at least path planning for the ego vehicle.