Automated Vehicle Situation Modeling for Real-Time Redundant Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automated vehicle systems have limited field of action and require significant hardware resources, making them inefficient for real-time operation and robustness in autonomous driving scenarios, especially at Level 5 autonomy.
Innovation Solution
A method and device utilizing a multi-agent approach with situation detection elements that computationally model surroundings data from various sensors, allowing for robust and efficient activation of driver assistance systems using overlapping models and self-assessment units to ensure real-time operation with hardware and software redundancy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional automated vehicle systems use comprehensive sensor data processing for all driving functions, then the field of action is expanded, but the hardware resources and computing effort increase significantly
Solution Approach 1:
The situation detection unit is divided into multiple independent situation detection elements, each responsible for specific detection tasks. This segmentation allows the system to process different aspects of the driving environment separately, reducing the computational burden on any single component while maintaining comprehensive coverage of the driving scene.
Solution Approach 2:
Multiple situation detection elements share common hardware resources and can be configured to perform different detection functions. The system uses a unified approach where the same hardware infrastructure supports various detection algorithms and scenarios, making the system versatile without proportionally increasing hardware requirements.
2Speed
If traditional automated vehicle systems process all sensor data in real-time, then the response speed is improved, but the computing effort and processing time increase
Solution Approach 1:
The system extracts only the essential information needed for each specific driving function from the comprehensive sensor data. Each situation detection element focuses on detecting specific situations relevant to its function, filtering out unnecessary data processing while maintaining real-time response capability.
Solution Approach 2:
The system performs partial processing of sensor data by allocating different processing depths to different situation detection elements based on their specific needs. Critical safety functions receive full processing attention, while less critical functions use optimized, lower-computation approaches, achieving real-time performance without excessive computing effort across the board.
3Device complexity
If automated vehicle systems use single sensor or single detection element, then the hardware resources are reduced, but the robustness and reliability decrease
Solution Approach 1:
Multiple situation detection elements are merged into a unified situation detection unit that shares common hardware resources. This merging approach allows the system to achieve redundancy and improved reliability through multiple detection elements working together, while conserving hardware resources by sharing sensors, processors, and memory across all elements.
Solution Approach 2:
The system changes the operational parameters of the shared hardware resources dynamically, allocating computing power and sensor access differently to each situation detection element based on current driving conditions and priorities. This allows the same hardware to support multiple detection functions with appropriate resource allocation, maintaining reliability without proportional hardware increase.
4Reliability
If automated vehicle systems implement comprehensive monitoring of all subsystems, then the reliability is improved, but the device complexity and control unit size increase
Solution Approach 1:
The situation detection elements act as intermediaries between the raw sensor data and the higher-level control decisions. Each element provides monitoring and assessment of its specific detection domain, filtering and preparing information for the decision-making unit. This intermediary layer distributes the monitoring complexity across multiple specialized elements rather than requiring a single monolithic control unit, reducing overall device complexity while maintaining comprehensive monitoring.
Data Source
AI summary
A method for operating an automated vehicle. The method includes: detecting surroundings of the vehicle; providing surroundings data of the detected surroundings; supplying the surroundings data to a situation detection unit including a defined number greater than one of situation detection elements; computationally modeling the surroundings of the vehicle with the aid of the situation detection elements; activating driver assistance systems using output data of the models of the situation detection elements; deciding, with the aid of a decision-making unit, which output data of the models of the situation detection elements are used for activating an actuator unit of the vehicle; and activating the actuator unit of the vehicle using the decided-upon output data.