AV Hazard Detection Using Trajectory Deviations of Nearby Traffic

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Autonomous vehicles (AVs) face challenges in detecting and responding to unknown driving hazards due to their lack of training on novel driving situations, which can impact safety for both the AV and surrounding road participants.

Innovation Solution

The AV utilizes an unknown driving hazard detection module to identify unusual behavior of nearby road participants through trajectory discrepancies, triggering safety responses such as remote assistance, safe stops, or mimicking surrounding traffic behavior, and logs data for future model training.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If the AV relies on pre-trained models for hazard detection, then the system has high reliability for known situations, but it cannot detect or respond to unknown driving hazards

Engineering Contradiction:

Improveability to detect unknown hazardsVSAvoidsafety response accuracy

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The system monitors the behavior of multiple road participants and compares actual trajectories against predicted trajectories. When discrepancies are detected, this feedback loop triggers a re-evaluation process that can identify unknown hazards and initiate appropriate safety responses, allowing the system to adapt to situations beyond its pre-trained knowledge

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system performs preliminary trajectory predictions for road participants based on their current behavior patterns. By having these predictions ready in advance and continuously comparing them with actual movements, the system can quickly identify deviations that indicate unknown hazards, enabling faster response times

Inventive Principle:

Principle #10Preliminary action

2Adaptability or versatility

If the AV monitors and evaluates behavior of multiple road participants to detect unusual patterns, then the ability to detect unknown hazards improves, but the computational complexity and processing time increase

Engineering Contradiction:

Improveunknown hazard detection capabilityVSAvoidcomputational processing complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system extracts only the critical behavior parameters of road participants (current position, velocity, and trajectory predictions) and focuses computational resources on analyzing deviations from predicted paths. By extracting only the essential data needed for hazard detection rather than processing all possible sensor data, the system reduces computational complexity while maintaining detection effectiveness

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The system creates simplified copies or representations of road participant behavior through trajectory predictions based on current motion patterns. These predictive models serve as lightweight approximations that can be quickly compared against actual behavior, reducing the computational burden of real-time analysis while maintaining the ability to detect unusual patterns

Inventive Principle:

Principle #26Copying

Data Source

PatentUS20250214624A1Unknown driving hazard detection and response system

Publication Date: 2025.07.03 GM CRUISE HOLDINGS LLC

AI summary

A system installed on an autonomous vehicle (AV) is described and includes an unknown driving hazard detection module configured to detect an unusual driving behavior of at least one road participant operating in proximity to the AV, wherein the unusual driving behavior comprises a deviation from a behavior of the at least one road participant as predicted by the AV; and an unknown driving hazard response module configured to cause an action to be performed in connection with the AV based on the detected unusual driving behavior.