Path Perception Diversity for Reliable Autonomous Path Planning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional autonomous vehicle systems rely on a single data source for path perception, which is impractical for real-time accuracy assessment and can lead to inaccurate information, especially in challenging scenarios, resulting in system failures and compromised passenger comfort and safety.
Innovation Solution
Implementing a diverse set of path perception inputs from deep neural networks, high-definition maps, and object traces to generate a more accurate and reliable understanding of the driving surface, enabling real-time assessment and redundancy in path computation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single data source is used for path perception, then device complexity is reduced, but reliability deteriorates due to inability to perform real-time accuracy assessment
Solution Approach 1:
The patent combines multiple path perception data sources including deep neural network outputs, high-definition map data, and object trace information into a unified ensemble assessment framework. This merging allows the system to cross-validate multiple independent sources simultaneously, enabling real-time reliability assessment without proportionally increasing system complexity
Solution Approach 2:
The ensemble assessment framework serves multiple functions: it validates path perception accuracy, identifies unreliable sources, and provides fallback mechanisms. This multi-functionality allows a single assessment system to handle various failure modes and scenarios, improving reliability without requiring separate specialized systems for each function
2Reliability
If multiple path perception inputs are used, then reliability is improved through redundancy, but device complexity increases
Solution Approach 1:
The patent segments the path perception system into distinct independent sources (DNN outputs, HD map modules, object trace generators) that can be individually assessed and validated. This segmentation allows the ensemble framework to evaluate each source's reliability separately and combine them systematically, managing complexity through modular organization rather than monolithic processing
Solution Approach 2:
The ensemble assessment framework implements continuous feedback by comparing multiple path perception sources in real-time, identifying discrepancies, and adjusting reliance on individual sources based on their agreement. This feedback mechanism automatically manages complexity by dynamically weighting sources based on their current reliability rather than requiring manual configuration of complex validation rules
3Speed
If a single path perception signal is relied upon, then processing speed is maintained, but measurement precision deteriorates due to inability to verify accuracy
Solution Approach 1:
The patent applies partial validation by assessing only the critical aspects of path perception accuracy through ensemble comparison rather than performing exhaustive verification of all possible error modes. This selective assessment maintains real-time processing speeds while providing sufficient precision verification for safe autonomous operation, avoiding the computational burden of complete accuracy verification
Data Source
AI summary
In various examples, a path perception ensemble is used to produce a more accurate and reliable understanding of a driving surface and/or a path there through. For example, an analysis of a plurality of path perception inputs provides testability and reliability for accurate and redundant lane mapping and/or path planning in real-time or near real-time. By incorporating a plurality of separate path perception computations, a means of metricizing path perception correctness, quality, and reliability is provided by analyzing whether and how much the individual path perception signals agree or disagree. By implementing this approach—where individual path perception inputs fail in almost independent ways—a system failure is less statistically likely. In addition, with diversity and redundancy in path perception, comfortable lane keeping on high curvature roads, under severe road conditions, and/or at complex intersections.