Sensor-Fusion Network for Inter-Vehicle Validation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Sensor systems in self-driving vehicles face reliability and safety issues due to changes in sensor functionality over time, environmental conditions, and unexpected events, leading to less reliable vehicle operations.
Innovation Solution
A vehicle sensor validation system utilizing a coordinating processor and sensor-fusion network across multiple vehicles to validate sensor data by comparing data from multiple sensors, employing a majority-voter algorithm to generate coordinated-validation data and adjust vehicle operations accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If sensor data is used directly without validation, then system complexity is reduced, but reliability of vehicle operation deteriorates due to sensor degradation and environmental factors
Solution Approach 1:
The patent merges sensor data from multiple vehicles into a unified sensor-fusion network, combining individual sensor readings to validate each other. This allows the system to improve reliability through cross-validation while managing complexity by processing fused data collectively rather than individually for each vehicle.
Solution Approach 2:
The coordinating processor acts as an intermediary that receives sensor data from multiple vehicles, performs validation operations, and distributes validated information back to the network. This mediator approach centralizes the complexity of validation logic, allowing individual vehicle systems to remain simpler while achieving high reliability through the intermediary's coordination.
2Reliability
If sensor functionality is monitored continuously, then reliability is improved, but energy consumption increases
Solution Approach 1:
The system implements periodic validation where sensors are monitored and cross-checked at intervals rather than continuously. The coordinating processor receives sensor data periodically from multiple vehicles and performs validation operations at these discrete time points, reducing energy consumption compared to continuous monitoring while maintaining reliability through regular checks.
Solution Approach 2:
Sensors validate each other through the sensor-fusion network without requiring continuous active monitoring of each individual sensor. The system uses the inherent data flow from multiple sensors to perform self-validation, where the normal operation of sensors provides the data needed for validation, eliminating the need for additional dedicated monitoring resources and reducing overall energy consumption.
3Measurement precision
If data from multiple sensors is processed for validation, then measurement precision is improved, but loss of time increases due to additional processing
Solution Approach 1:
The system performs preliminary data fusion and validation operations in advance before critical decisions are needed. The coordinating processor continuously fuses sensor data from multiple vehicles and pre-validates sensor readings, so that when validation is needed for autonomous driving decisions, the work is already completed or nearly completed, reducing the time loss during critical moments.
Solution Approach 2:
The patent combines sensor data processing with the normal data exchange operations in the vehicle network. Instead of adding separate validation processing steps, the system merges validation operations with the existing sensor-fusion data aggregation process, allowing multiple vehicles to contribute data that is simultaneously used for both fusion and validation purposes, thereby improving measurement precision without proportionally increasing processing time.
Data Source
AI summary
A system and method of sensor validation utilizing a sensor-fusion network. The sensor-fusion network may comprise a number of sensors associated with one or more vehicles having autonomous or partially autonomous driving functions.