Vehicle FPGA Map Tile Fusion for Real-Time Object Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current high-definition (HD) electronic navigation maps for autonomous vehicles struggle to reflect real-time road dynamics, such as moving objects, due to limited sensor capabilities and inefficient network resource usage, and lack scalability and ubiquitous coverage, especially in areas with low vehicle density or network failures.

Innovation Solution

A system that employs in-vehicle hardware for real-time sensor fusion and map tile fusion, integrating moving object sensor data with HD map tiles, and uses RSUs for efficient wireless network resource management and local data distribution, including minimal viable map tiles for navigation in emergency cases.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If sensor data is integrated with HD map tiles to reflect real-time road dynamics, then situational awareness is enhanced, but device complexity increases

Engineering Contradiction:
Improvesituational awarenessVSAvoiddevice complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent introduces an intermediary processing layer that fuses sensor data with HD map tiles through standardized interfaces. This mediator component handles the complexity of real-time data integration, allowing the system to reflect road dynamics without requiring complete system redesign. The intermediary layer abstracts the complexity while maintaining reliability through structured data fusion protocols.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system segments the map data into tiles that can be independently updated and processed. By dividing the HD map into manageable tile segments, the system can integrate sensor data locally for specific road segments without processing entire maps, reducing computational complexity while maintaining real-time situational awareness for relevant areas.

Inventive Principle:
Principle #1Segmentation

2Speed

If map tiles are distributed through wireless networks, then update speed improves, but network resource usage increases

Engineering Contradiction:
Improveupdate speedVSAvoidnetwork resource usage
Core Design Contradiction:
SpeedVSLoss of energy

Solution Approach 1:

The patent implements local quality by distributing map tile updates selectively through wireless networks only where and when needed. Instead of broadcasting entire map updates network-wide, the system identifies specific geographic areas requiring updates and transmits only those relevant tiles to vehicles in those locations, optimizing network resource usage while maintaining fast update speeds for affected regions.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system performs preliminary actions by pre-processing and segmenting map data into ready-to-distribute tiles before network transmission. This preliminary segmentation and preparation of map tiles allows for efficient network delivery, reducing the actual transmission time and network resource consumption during real-time updates.

Inventive Principle:
Principle #10Preliminary action

3Quantity of substance

If HD maps are segmented into tiles, then storage requirements decrease, but data integration complexity increases

Engineering Contradiction:
Improvestorage requirementsVSAvoiddata integration complexity
Core Design Contradiction:
Quantity of substanceVSDevice complexity

Solution Approach 1:

The patent applies the nested doll principle by organizing map tiles in a hierarchical structure where smaller detail tiles are nested within larger regional map contexts. This nested organization allows the system to store only relevant detailed tiles for current locations while maintaining access to broader geographic contexts, reducing overall storage requirements while simplifying integration through the hierarchical framework.

Inventive Principle:
Principle #7Nested doll (Nesting)

Data Source

PatentUS11879749B2Enhanced high definition maps for a vehicle
Publication Date: 2024.01.23 INTEL CORP
  • US11879749B2 patent drawing
  • US11879749B2 patent drawing
  • US11879749B2 patent drawing

AI summary

System and techniques for enhanced electronic navigation maps for a vehicle are described herein. A set of map tiles may be received at a vehicle component from a remote entity. Sensor derived data that has a locality corresponding to a map tile in the set of map tiles may be obtained. A field-programmable gate array of the vehicle may then be invoked to combine the sensor derived data and the map tile to create a modified map tile. The modified map tile may be communicated to a control system of the vehicle.