Autonomous Vehicle Trajectory Diagnostics for Prediction and Map Errors

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Autonomous vehicles face challenges in planning optimal trajectories due to sub-optimal or inaccurate scene understanding, which can lead to unsafe operations, necessitating effective detection and remediation of prediction and map errors.

Innovation Solution

A method and apparatus for real-time map and prediction diagnostics that detect road users, calculate average prediction displacement error (APDE) and average map displacement error (AMDE), and determine accuracy based on these errors to assess scene understanding and adjust trajectory planning accordingly.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If autonomous vehicles use scene understanding to plan trajectories, then navigation efficiency is improved, but prediction accuracy and map accuracy deteriorate due to sub-optimal scene understanding

Engineering Contradiction:

Improvenavigation efficiencyVSAvoidprediction accuracy

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The system continuously monitors actual trajectories of road users and compares them with predicted trajectories from scene understanding models. Prediction displacement errors are calculated and used to provide feedback for detecting prediction errors and adjusting the scene understanding models, thereby improving prediction accuracy while maintaining navigation efficiency

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system pre-calculates predicted trajectories for multiple road users before actual conflicts occur. By storing both predicted and actual trajectory data in advance, the system can perform real-time error detection and diagnosis without compromising navigation decision-making speed

Inventive Principle:

Principle #10Preliminary action

2Measurement precision

If autonomous vehicles use high-definition maps for navigation, then route accuracy is improved, but map error detection capability deteriorates without dedicated diagnostic systems

Engineering Contradiction:

Improveroute accuracyVSAvoidmap error detection

Core Design Contradiction:

Measurement precisionVSDifficulty of detecting and measuring

Solution Approach 1:

The system compares actual road user trajectories with mapped drivelines from high-definition maps. Map displacement errors are calculated by analyzing deviations between actual paths and mapped paths, providing continuous feedback for detecting map inaccuracies while maintaining route navigation accuracy

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system uses actual road user trajectories as an intermediary to indirectly detect map errors. Instead of directly validating map data, the system observes how road users actually traverse the environment and uses these observations to infer map inaccuracies through displacement error analysis

Inventive Principle:

Principle #24Intermediary (Mediator)

3Reliability

If autonomous vehicles implement real-time prediction and map diagnostics, then safety is improved, but system complexity increases due to additional data processing requirements

Engineering Contradiction:

ImprovesafetyVSAvoidsystem complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The diagnostic functionality is extracted as a separate module that independently processes trajectory data. By separating prediction error detection and map error detection into distinct diagnostic subsystems, the main navigation system maintains its original complexity while the diagnostic module handles the additional computational requirements for safety monitoring

Inventive Principle:

Principle #2Taking out (Extraction)

4Measurement precision

If autonomous vehicles store and analyze multiple trajectory data points, then prediction error detection accuracy is improved, but data processing time increases

Engineering Contradiction:

Improveprediction error detection accuracyVSAvoiddata processing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system calculates prediction displacement errors using a selected subset of trajectory data points rather than processing all available data. By using at least two corresponding predicted data points and one actual data point to compute average prediction displacement error, the system achieves sufficient detection accuracy while reducing computational time compared to exhaustive analysis of all trajectory points

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS12077177B2Autonomous vehicle control and map accuracy determination based on predicted and actual trajectories of surrounding objects and vehicles

Publication Date: 2024.09.03 NISSAN MOTOR CO LTD

AI summary

Detecting prediction errors includes detecting a road user; determining respective predicted data for the road user; storing, in a data structure, the respective predicted data; storing, in the data structure, actual data of the road user; obtaining an average prediction displacement error using at least one of the actual data and at least two corresponding respective predicted data; and determining a prediction accuracy based on the average prediction displacement error. Detecting map errors includes detecting a road user; storing, in a data structure, actual data of the road user; storing, in the data structure, map data corresponding to the actual data; obtaining an average map displacement error based on a comparison of at least some of the actual data and corresponding at least some map data; and determining a map accuracy based on the average map displacement error.