Vehicle Sensor Coordination Using Neural Network Subregion Targeting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sensor systems in vehicles process data independently, lacking an efficient method to integrate and enhance monitoring capabilities across different types of sensors, leading to incomplete and inefficient surveillance of the vehicle's surroundings.
Innovation Solution
A method utilizing a neural network to generate a control signal for a second sensor system based on data from a first sensor system, enabling improved monitoring of specific subregions by adjusting the second sensor's parameters, such as angle and distance, without requiring expert knowledge.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple sensor systems process data independently, then each sensor system can be optimized for its specific function, but the overall monitoring capability and integration efficiency deteriorate
Solution Approach 1:
The patent merges multiple sensor systems (video camera, radar, lidar) into a unified monitoring framework where a central processing unit integrates data from all sensors. This combining approach improves overall monitoring capability by leveraging the strengths of each sensor type while maintaining systematic coordination, directly resolving the contradiction between reliability and complexity.
Solution Approach 2:
The processing unit serves multiple functions: it processes data from different sensor types, generates control signals for controllable sensors, and coordinates monitoring across various regions. This multi-functionality allows the system to achieve comprehensive monitoring capability without proportionally increasing system complexity, as a single unit handles diverse tasks.
2Area of stationary object
If the entire 360° field is monitored continuously, then complete surveillance coverage is achieved, but the detection precision and attention to critical regions deteriorate
Solution Approach 1:
The patent implements local quality by directing controllable sensor systems to focus on specific subregions of interest rather than uniformly monitoring the entire 360° field. The processing unit identifies critical regions and allocates monitoring resources preferentially to those areas, achieving high detection precision in critical zones while maintaining overall coverage through other sensor systems.
Solution Approach 2:
The system dynamically adjusts monitoring focus based on detected objects and situations. Controllable sensor systems redirect their attention to relevant regions as conditions change, allowing the system to maintain precise detection in active monitoring zones while preserving the ability to cover the entire area through coordinated sensor deployment.
3Loss of information
If all sensor systems operate at full capacity simultaneously, then comprehensive data collection is achieved, but the energy consumption and system cost deteriorate
Solution Approach 1:
The patent applies partial action by activating sensor systems and adjusting their operational parameters based on actual monitoring needs rather than running all sensors at full capacity continuously. The processing unit determines which sensors should be active and at what intensity, collecting sufficient data for comprehensive monitoring while avoiding unnecessary energy expenditure from excessive sensor operation.
Solution Approach 2:
The system employs periodic action through selective activation of sensor systems based on detected conditions. Rather than continuous full-capacity operation, sensors are activated periodically or on-demand based on the presence of objects of interest, allowing the system to maintain data completeness while significantly reducing average energy consumption compared to continuous operation.
Data Source
AI summary
A method for monitoring surroundings of a first sensor system. The method includes: providing a temporal sequence of data of the first sensor system for monitoring the surroundings; generating an input tensor including the temporal sequence of data of the first sensor system, for a trained neural network; the neural network being configured and trained to identify, on the basis of the input tensor, at least one subregion of the surroundings, in order to improve the monitoring of the surroundings with the aid of a second sensor system; generating a control signal for the second sensor system with the aid of an output signal of the trained neural network, in order to improve the monitoring of the surroundings in the at least one subregion.