Vehicle Environmental Model Updates Using Confidence-Based Delay
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous driving vehicles face challenges in identifying and classifying objects, leading to the need for teleoperated driving support, which can result in excessive data sharing and signaling overhead, especially when sensor malfunctions or uncertainties occur.
Innovation Solution
A method for updating environmental models at vehicles by determining confidence improvements in object identification using independent sensor data and prioritizing message transmission based on confidence levels, reducing redundant information sharing and signaling overhead.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If sensor data is shared between vehicles to improve environmental model accuracy, then object identification reliability is improved, but communication overhead and data transmission load increase
Solution Approach 1:
The patent extracts only the essential confidence information and object parameters from the environmental model, transmitting only this condensed data between vehicles rather than sharing complete sensor datasets. This extraction principle reduces communication overhead while maintaining the ability to improve object identification reliability through inter-vehicle information exchange.
Solution Approach 2:
The patent transforms the environmental model data into a different parameter representation by using confidence levels and selective parameter transmission. Instead of transmitting raw sensor data or complete environmental models, the system changes the data parameters to confidence scores and essential object attributes, reducing transmission volume while preserving reliability improvement capability.
2Loss of time
If all vehicles transmit environmental model updates simultaneously, then information freshness is improved, but communication channel overload increases
Solution Approach 1:
The patent implements periodic transmission based on confidence thresholds rather than continuous or simultaneous transmission. Vehicles transmit environmental model updates periodically when confidence levels change significantly, which maintains information freshness while avoiding communication channel overload from simultaneous transmissions.
Solution Approach 2:
The patent makes the transmission timing dynamic by adjusting it based on confidence level changes and traffic conditions. Instead of fixed periodic transmission or simultaneous transmission, the system dynamically determines when transmission is necessary based on environmental model confidence changes, optimizing both information freshness and communication efficiency.
3Reliability
If teleoperated driving support is activated for sensor malfunctions, then safety is improved, but response time and communication latency increase
Solution Approach 1:
The patent performs preliminary actions by pre-establishing confidence thresholds and communication protocols before sensor malfunctions occur. When malfunctions happen, the system can immediately activate teleoperated support using pre-configured parameters, reducing response time while maintaining safety through pre-validated communication channels and decision criteria.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Embodiments provide a method, a computer program, an apparatus, a vehicle, and a traffic entity for updating an environmental model at a vehicle. The method (10) for a traffic entity (100) and for updating a first environmental model at a vehicle (200) comprises receiving (12) information related to the first environmental model from the vehicle (200). The first environmental model comprises at least first information on an object in an environment of the vehicle (200) and the first environmental model comprising confidence information related to the first information. The method (10) comprises obtaining (14) information related to a second environmental model of the vehicle (200) at the traffic entity (100). The second environmental model comprising at least second information on the object in the environment of the vehicle (200) and the second environmental model comprising confidence information related to the second information. The method (10) comprises determining (16) information related to a confidence improvement of the first information based on the second information, determining (18) a delay based on the information related to the confidence improvement, and transmitting (20) the second information to the vehicle (200) based on the delay.