Autonomous Vehicle Reference Lines from Radar Arc Curves

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for generating a reference line for autonomous driving vehicles are inefficient when camera and digital map data are unavailable due to hardware or software failures, lacking effective alternatives for trajectory planning.

Innovation Solution

The method involves using sensor data from multiple sensors like cameras and RADAR devices to detect objects, generate arc curves, and select the object with the smallest curvature to create a reference line, allowing for trajectory planning without relying on camera or digital map data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If camera and digital map based methods are used to generate reference line, then measurement precision and reliability are improved under normal conditions, but the system becomes vulnerable to hardware or software failures

Engineering Contradiction:
Improvereference line generation reliabilityVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent introduces an intermediary method using radar sensor data and arc curve fitting as a mediator between the primary camera/digital map system and the fallback mechanism. When camera or digital map data becomes unavailable, the system uses radar-detected objects as intermediate reference points to generate arc curves, which serve as substitutes for the original reference line source, maintaining system operation without direct dependency on the failed component

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system dynamically changes operational parameters by switching from camera-based visual data to radar-based electromagnetic wave data when failures occur. The patent modifies the data source parameter from optical sensors to electromagnetic sensors, and changes the processing method from direct digital map matching to arc curve fitting based on detected objects, allowing the system to adapt to different operational conditions

Inventive Principle:
Principle #35Parameter changes

2Reliability

If multiple sensors and arc curve calculations are implemented, then tolerance to hardware or software failures is improved, but computational complexity and processing time increase

Engineering Contradiction:
Improvefailure toleranceVSAvoidcomputational complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent extracts only the essential computational elements needed for reference line generation when primary systems fail. Instead of processing all sensor data comprehensively, the system extracts only radar-detected object positions and uses a simplified arc curve fitting algorithm focused on generating a single reference line, reducing computational burden while maintaining failure tolerance

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The system employs a disposable computational approach by using temporary arc curve representations based on detected objects. Rather than maintaining complex, long-term models of the environment, the system creates simple, computationally inexpensive arc curves from current radar data that can be quickly generated and discarded, reducing overall computational complexity

Inventive Principle:
Principle #27Cheap short-living objects (Disposable)

Data Source

PatentEP3697659B1Method and system for generating reference lines for autonomous driving vehicles
Publication Date: 2023.11.22 BAIDU COM TIMES TECH (BEIJING) CO LTD
  • EP3697659B1 patent drawingFigure 1
  • EP3697659B1 patent drawingFigure 2
  • EP3697659B1 patent drawingFigure 3A

AI summary

A computer-implemented method for operating an autonomous driving vehicle is provided, the method comprising: perceiving a driving environment surrounding an autonomous driving vehicle (ADV), including identifying one or more objects based on sensor data obtained from a plurality of sensors of the ADV; for each of the identified objects, generating an arc curve connecting a current location of the ADV and the object, and calculating a curvature of the arc curve associated with the object; selecting one of the objects associated with an arc curve that satisfies a predetermined condition; and generating a reference line from the current location of the ADV to the selected object, wherein the reference line is utilized to generate a trajectory to drive the ADV. A non-transitory machine-readable medium and a data processing system is also provided.