Vehicle Surrounding Map Fusion for Dead Angle Coverage

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Autonomous driving systems face challenges in creating precise condition recognition maps due to failures in outside recognition sensors, leading to incomplete maps and potential collisions, especially in scenarios with dead angles caused by sensor malfunctions or environmental obstructions.

Innovation Solution

The system requests and merges information from external apparatuses, such as neighboring vehicles, to complement dead angles in the map, using V2X communication to acquire necessary data and enhance map accuracy, thereby ensuring safe navigation and preventing accidents.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If map information is created using only on-vehicle sensors, then the system maintains independence and real-time accuracy, but dead angles and sensor failures create incomplete map coverage

Engineering Contradiction:
Improvemap accuracyVSAvoiddead angle coverage
Core Design Contradiction:
Measurement precisionVSLoss of information

Solution Approach 1:

The patent merges map information from multiple external apparatuses (neighboring vehicles, infrastructure) with on-vehicle sensor data to create a comprehensive surrounding area map. This combination fills dead angles and compensates for sensor failures by integrating diverse data sources that provide complementary coverage perspectives.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

External apparatuses act as intermediaries that provide additional map information about areas not directly observable by on-vehicle sensors. These intermediaries bridge the information gap in dead angles by transmitting map data captured from different spatial positions and perspectives.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Loss of information

If the system requests map information from multiple external apparatuses, then dead angles are complemented, but communication overhead and information processing complexity increase

Engineering Contradiction:
Improvedead angle complementationVSAvoidcommunication system complexity
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The system selectively requests map information from external apparatuses only when dead angles are detected or when sensor failures occur, rather than continuously exchanging data. This partial action approach complements information only when necessary, reducing unnecessary communication overhead while still achieving dead angle coverage.

Inventive Principle:
Principle #16Partial or excessive action

Solution Approach 2:

External apparatuses pre-create and store map information in advance, so that when requests are made, the data is already available for immediate transmission. This preliminary preparation reduces real-time processing complexity during actual dead angle complementation operations.

Inventive Principle:
Principle #10Preliminary action

3Reliability

If sensor information is continuously monitored for failures, then system reliability is maintained, but computation and processing loads increase

Engineering Contradiction:
Improvesensor failure detectionVSAvoidcomputation energy consumption
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The system implements feedback mechanisms where map creation results are continuously evaluated for completeness and consistency. When discrepancies or gaps are detected in the generated map, this triggers targeted sensor functionality checks and external information requests, rather than continuous monitoring of all sensors.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

Instead of continuously monitoring all sensor functions, the system performs partial monitoring focused on detecting map creation failures and inconsistencies. This selective monitoring approach maintains reliability by detecting when map information is insufficient while reducing computation energy compared to exhaustive sensor monitoring.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentUS11965754B2Information processing apparatus, information processing method, and mobile body apparatus
Publication Date: 2024.04.23 SONY SEMICON SOLUTIONS CORP
  • US11965754B2 patent drawing
  • US11965754B2 patent drawing
  • US11965754B2 patent drawing

AI summary

Provided is an information processing apparatus that creates map information on the basis of sensor information obtained by an on-vehicle sensor. The information processing apparatus includes a creation section that creates a map of a surrounding area of a mobile body on the basis of sensor information acquired by one or more sensors mounted on the mobile body, a request section that issues an information request to an external apparatus on the basis of a state of the map created by the creation section, and a merge section that merges information acquired by the request section from the external apparatus with the created map. The request section issues an information request to the external apparatus on the basis of a condition of a dead angle included in the map created by the creation section.