Autonomous Vehicle Navigation With On-Demand HD Map Fragments
Find Innovative SolutionsGenerate 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
Engineering Contradiction Analysis
1Measurement precision
If high definition map is transmitted to vehicle apparatus, then navigation accuracy is improved, but bandwidth consumption increases significantly
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.
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.
2Reliability
If high definition map is stored in vehicle apparatus, then autonomous navigation capability is improved, but memory usage increases significantly
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.
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.
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
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.
Data Source
Figure 1
Figure 2A~2B
Figure 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.