Autonomous Vehicle Test Run Retrieval for Salient Planner Scenarios

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for testing autonomous vehicle systems struggle to efficiently identify salient scenarios, as determining the saliency of test runs in real or simulated environments is challenging due to the large volume of data and the difficulty in identifying which scenarios provide meaningful assessments of performance.

Innovation Solution

A computer system that processes run data to identify salient scenarios by detecting relevant circumstances, such as challenger agent behaviors, and applies predefined driving action assessment rules to determine the permissibility of actions, allowing for the retrieval and visualization of salient scenarios.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If multiple different trajectory planners are used to ensure comprehensive testing coverage, then the reliability of trajectory planner testing is improved, but the complexity of the testing system increases

Engineering Contradiction:
Improvetesting coverageVSAvoidtesting system
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The testing system is segmented by creating separate executable files for each trajectory planner type (e.g., A*, Dijkstra, BFS, DFS, Greedy Best-First, Uniform Cost, A* with heuristic, Dijkstra with heuristic). Each planner is tested independently through its own executable, allowing comprehensive coverage while maintaining manageable complexity through modular organization.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Instead of implementing multiple different trajectory planner algorithms within a single testing system, the approach copies the entire testing framework multiple times - once for each planner type. Each copy is customized with the specific planner implementation, creating identical testing structures that can be independently executed and managed.

Inventive Principle:
Principle #26Copying

2Measurement precision

If all trajectory planners are tested with every possible map and robot configuration, then the measurement precision of testing is improved, but the time required for testing increases

Engineering Contradiction:
Improvetesting thoroughnessVSAvoidtesting duration
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The comprehensive test suite is segmented into multiple independent executable files, each dedicated to testing a specific trajectory planner. This segmentation allows parallel execution of tests across different planners, reducing the total time required while maintaining thorough testing coverage for each individual planner.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The testing framework is pre-configured with all necessary test cases, map data, and robot configurations embedded within each executable file before execution. This preliminary preparation eliminates the need for runtime configuration and allows tests to run immediately with full thoroughness, reducing overall testing time.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP4533269B1Identifying salient test runs involving mobile robot trajectory planners
Publication Date: 2026.04.29 FIVE AI LTD
  • EP4533269B1 patent drawingFigure 1A
  • EP4533269B1 patent drawingFigure 1B
  • EP4533269B1 patent drawingFigure 1C

AI summary

The disclosure provides systems and methods for identifying salient test runs involving an autonomous vehicle system. A processor receives sets of run data, each set representative of a driving scenario. For each set, an output set is generated, the output set comprising time-indexed events generated in response to a detected behaviour of at least one challenger agent, and a sequence of decision indicators indicating whether a driving action by an ego agent would be permissible. A data retrieval component is coupled to a results database and retrieves output sets based on the time-indexed events and the sequence of decision indicators. The processor generates the sequence of decision indicators by generating a planned trajectory of the ego agent, and determining whether the predefined driving action by the ego agent would be permissible.