Autonomous Vehicle ODD Identification via Bounded-Risk Geographic Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for defining the Operational Design Domain (ODD) of autonomous driving systems rely heavily on safety drivers, who may not be adequately prepared for potential failures, leading to increased risk and burnout, and lack a systematic way to distinguish safe and unsafe areas, resulting in potentially unsafe road testing.
Innovation Solution
A system and method that objectively define the ODD based on the capabilities of the autonomous driving system, using a geographic dataset to identify bounded-risk areas, allowing for real-time alerts to operators and systematic updates, thereby reducing reliance on safety drivers and ensuring bounded risk thresholds.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If safety drivers are used to monitor and intervene during ADS testing, then operator awareness of system capabilities is maintained, but operator burnout and loss of focus occur over time
Solution Approach 1:
The ADS system performs self-evaluation of its performance and capabilities through automated testing and data collection. The system generates its own performance metrics and capability assessments without requiring continuous human monitoring, allowing the system to serve its own safety evaluation needs.
Solution Approach 2:
The patent replaces the mechanical system of human safety drivers with an automated electronic system that collects sensor data, processes performance metrics, and determines ODD boundaries algorithmically. This substitution eliminates human fatigue while maintaining objective safety assessment through data-driven methods.
2Reliability
If safety drivers are required to be fully alert at all times, then immediate intervention capability is ensured, but reaction time is increased due to lack of advance warning
Solution Approach 1:
The system performs preliminary evaluation of ADS performance through automated testing and data collection before deployment. By pre-assessing system capabilities and identifying safe operational boundaries through algorithmic analysis, the system prepares safety assessments in advance rather than relying on real-time human reaction.
Solution Approach 2:
The patent implements continuous feedback loops where sensor data is collected, processed, and used to update performance metrics and ODD definitions. This automated feedback system provides immediate system response to performance changes without requiring human interpretation or reaction time.
3Productivity
If the entire geo-fenced area is claimed as ODD after testing, then deployment coverage is maximized, but unsafe areas are included due to lack of systematic distinction
Solution Approach 1:
The patent segments the geo-fenced testing area into distinct zones based on automated performance evaluation. Rather than treating the entire area as uniform ODD, the system divides it into safe and unsafe portions using algorithmic analysis of sensor data and performance metrics, allowing selective deployment in validated safe zones.
Solution Approach 2:
The system uses parameter-based thresholds to objectively define ODD boundaries. By establishing quantitative performance criteria and using automated analysis to measure actual ADS performance against these parameters, the system dynamically determines safe operational areas based on measured capabilities rather than arbitrary geographic boundaries.
4Adaptability or versatility
If safety drivers are continually briefed on ADS capability changes, then operator knowledge is kept current, but operator workload and complexity increase
Solution Approach 1:
The ADS system automatically tracks and communicates its own capability changes through automated reporting and data logging. The system serves its own information needs by maintaining performance records and capability metadata, eliminating the need for manual briefings while keeping operational data current and accessible.
Solution Approach 2:
The patent replaces the manual process of briefing safety drivers with automated electronic communication systems. Capability updates are transmitted digitally to relevant systems and personnel, reducing human workload while ensuring accurate, consistent information distribution without manual intervention.
Data Source
AI summary
A system, method and processor readable medium for identifying an operational design domain (ODD) for operation of an autonomous driving system (ADS). In one aspect, a proposed map is used to generate a geographic dataset. Performance of the ADS is evaluated against the geographic dataset under a range of environment& conditions, thereby identifying a bounded-risk portion of a proposed condition space defined by the proposed map and the range of environmental conditions. The operational design domain is identified based on the bounded-risk portion of the proposed condition space. The ODD can be updated as additional data is received. When a vehicle using the ADS is likely to leave the ODD, an operator can be alerted.