Autonomous Vehicle Navigation With On-Demand HD Map Fragments

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Autonomous vehicle navigation requires high definition maps that are bandwidth-intensive and memory-heavy, with details valid for only a limited time, posing a challenge in efficient data transmission and storage.

Innovation Solution

Utilizing non-connected and non-contiguous map fragments that are transmitted only when needed, allowing vehicles to operate in sensor-based mode where sensors are sufficient and switch to map-based mode where sensors are insufficient, reducing bandwidth and memory usage.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If high definition map is transmitted to vehicle apparatus, then navigation accuracy is improved, but bandwidth consumption increases significantly

Engineering Contradiction:
Improvenavigation accuracyVSAvoidbandwidth consumption
Core Design Contradiction:
Measurement precisionVSLoss of energy

Solution Approach 1:

The high definition map is divided into multiple map fragments corresponding to different geographic areas. The vehicle apparatus receives only the specific map fragment needed for its current location, rather than the complete high definition map. This segmentation approach maintains navigation accuracy in the relevant area while significantly reducing bandwidth consumption by transmitting only necessary map data portions.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Different map data is provided for different geographic areas along the vehicle's route. The system identifies which specific areas require high definition map data and transmits only those localized portions. This ensures navigation accuracy is maintained where needed while avoiding transmission of redundant map data for areas where sensors suffice, thereby reducing overall bandwidth consumption.

Inventive Principle:
Principle #3Local quality

2Reliability

If high definition map is stored in vehicle apparatus, then autonomous navigation capability is improved, but memory usage increases significantly

Engineering Contradiction:
Improveautonomous navigation capabilityVSAvoidmemory usage
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The high definition map is segmented into multiple map fragments that are stored separately in the vehicle apparatus. Only the map fragment corresponding to the vehicle's current operational area is loaded into memory, rather than storing the entire high definition map. This maintains autonomous navigation capability in the relevant area while significantly reducing memory usage requirements.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system proactively determines which map fragments will be needed based on the vehicle's route and sensor capabilities, and pre-loads only those specific fragments into memory before they are needed. This ensures navigation capability is ready when needed while avoiding the memory overhead of storing complete high definition maps that may never be used.

Inventive Principle:
Principle #10Preliminary action

3Loss of energy

If sensor-based navigation mode is used, then bandwidth and memory usage are reduced, but navigation reliability decreases in areas where sensors are insufficient

Engineering Contradiction:
Improvebandwidth and memory usageVSAvoidnavigation reliability
Core Design Contradiction:
Loss of energyVSReliability

Solution Approach 1:

The system dynamically switches between sensor-based navigation mode and map-based navigation mode depending on the vehicle's location and sensor performance. In areas where sensors provide sufficient data, the system uses sensor-based mode to minimize bandwidth and memory usage. When sensors become insufficient, the system transitions to map-based mode using pre-loaded map fragments, thereby maintaining navigation reliability while optimizing resource consumption across different operational contexts.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP4249851B1Autonomous vehicle navigation using map fragments
Publication Date: 2025.11.26 HERE GLOBAL BV
  • EP4249851B1 patent drawingFigure 1
  • EP4249851B1 patent drawingFigure 2A~2B
  • EP4249851B1 patent drawingFigure 3

AI summary

An apparatus autonomously operates at least one system of a vehicle as the vehicle navigates along a route through at least a portion of a traversable network in a sensor-based navigation mode. The apparatus obtains a map fragment corresponding to an area along the route; and, responsive to determining that the vehicle is located within the area, autonomously operates the at least one system of the vehicle as the vehicle navigates a portion of the route within the area in a map-based navigation mode using map data of the map fragment. Responsive to determining that the vehicle has exited the area, the apparatus autonomously operates the at least one system of the vehicle as the vehicle continues to navigate along the route in the sensor-based navigation mode. The map fragment is a portion of a high definition map configured to enable map-based autonomous vehicle navigation/control.