Vehicle Target Confidence Scoring via Sensor Weighting

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current perception technologies for vehicles lack robust and reliable object detection, particularly in varying light conditions and environmental impacts, as they do not adequately account for sensor intrinsic characteristics and environmental factors.

Innovation Solution

A method to assign a confidence score to detected targets based on intrinsic sensor characteristics and environmental conditions, including type, field of action, atmospheric conditions, and target properties, using weighting coefficients to update the confidence score over time.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If multiple sensors are used to detect targets, then the quantity of detection information increases, but the complexity of processing and integrating this information increases

Engineering Contradiction:
Improvequantity of detection informationVSAvoidcomplexity of processing and integrating information
Core Design Contradiction:
Quantity of substanceVSDevice complexity

Solution Approach 1:

The patent segments the sensor system into multiple independent detection modules, each responsible for specific sensor types (camera, radar, lidar). Each module processes its sensor data independently through detection, weighting, and confidence scoring, then integrates results at a higher level. This segmentation reduces overall processing complexity while maintaining comprehensive multi-sensor detection capability.

Inventive Principle:
Principle #1Segmentation

2Productivity

If sensor data is processed without considering environmental factors, then the processing speed increases, but the reliability of target detection decreases

Engineering Contradiction:
Improveprocessing speedVSAvoidreliability of target detection
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent performs preliminary actions by pre-defining environmental condition categories (good, moderate, poor lighting, weather conditions) and pre-establishing weighting coefficients for different sensor types under these conditions. During real-time processing, the system only needs to identify the current environmental category and apply the corresponding pre-computed weights, avoiding complex real-time environmental analysis while maintaining high detection reliability.

Inventive Principle:
Principle #10Preliminary action

3Measurement precision

If confidence scores are updated continuously, then the accuracy of target recognition improves, but the computational load increases

Engineering Contradiction:
Improveaccuracy of target recognitionVSAvoidcomputational load
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent implements dynamic confidence scoring where the weighting coefficients are adjusted based on real-time environmental conditions and sensor performance. The system dynamically modifies the importance of different sensors and detection results according to current conditions (e.g., reducing camera reliability in poor lighting), allowing continuous accuracy improvement while adapting computational requirements to actual operational needs rather than using fixed high computational load.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP3915040B1Method for determining a confidence score for a target in surroundings of a vehicle
Publication Date: 2025.01.01 AMPERE SAS
  • EP3915040B1 patent drawingFigure 1~2
  • EP3915040B1 patent drawingFigure 3~4a
  • EP3915040B1 patent drawingFigure 4b~4c

AI summary

The invention relates to a method for determining a confidence score for a target, comprising the following steps: - a detection step in which a sensor (21, 22) detects a target to be tailed (3) and a computer (4) initializes a confidence score for said target to be tailed (3); - an acquisition step in which an active sensor (2) acquires data in order to detect targets (3,10) in the surroundings; - a weighting step in which the computer (4) assigns a weighting coefficient to the target to be tailed (3) on the basis of the detection or lack of detection by the active sensor (2) of the target to be tailed (3), weighted on the basis of features (50) of the active sensor (2) and of the surroundings (21) of the vehicle (1); - a scoring step in which the computer (4) updates the confidence score for the target to be tailed (3).