Motor Vehicle Sensor Selection for Computing Load Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current motor vehicle systems for detecting surroundings utilize sensor data inefficiently, as they either display data or perform object recognition separately, exceeding the computing power of existing units due to the vast amount of data, and fail to optimally combine all available information.
Innovation Solution
Implement a motor vehicle with multiple sensor devices, including a bird view or mirror replacement system, that uses a control device to selectively prioritize and utilize sensor devices based on operating conditions, allowing for efficient data evaluation by centralizing or decentralizing processing units to optimize computing capacity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If all sensor devices are used for object recognition continuously, then object recognition accuracy is improved, but computing power is exceeded
Solution Approach 1:
The system dynamically adjusts which sensor devices are used for object recognition based on the current operating condition. The control device selects different subsets of sensor devices depending on whether the vehicle is moving forward, reversing, or stationary, thereby adapting the computing load to the actual situational requirements while maintaining sufficient object recognition accuracy.
Solution Approach 2:
Instead of continuously using all sensor devices, the system employs partial action by selectively activating only the necessary subset of sensor devices for object recognition based on the current operating condition. This reduces the computing load while still achieving adequate object detection for the specific situation.
2Device complexity
If sensor data is displayed without object recognition, then system complexity is reduced, but information utilization is suboptimal
Solution Approach 1:
The system segments the processing of sensor data into two distinct paths: one for display purposes and another for object recognition. The control device determines which sensor devices feed into object recognition based on operating conditions, while other sensor data can be displayed without undergoing complex recognition processing, thus optimizing both complexity and information utilization.
3Reliability
If rear-oriented sensor devices are used during reversing, then reversing safety is improved, but front sensor devices become redundant
Solution Approach 1:
The system implements multi-functionality by enabling front-oriented sensor devices to serve dual purposes: their primary function for forward driving and an secondary function for object recognition during reversing operations. The control device activates these front sensors for object recognition when reversing, allowing them to contribute to situational awareness beyond their primary orientation.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention relates to a motor vehicle (10), preferably a commercial vehicle, with a plurality of sensor devices for environmental sensing. The plurality of sensor devices shall include at least one image acquisition device (1a, 1b) of a bird's-eye view system or a mirror replacement system. In addition, the motor vehicle (10) according to the invention comprises an evaluation unit (2) configured to perform object recognition based on sensor data from the plurality of sensor devices. Furthermore, the motor vehicle (10) has a control unit (3) configured to select a subset of the plurality of sensor devices, based on the existence of a predetermined operating condition, on which object recognition is based.This advantageously achieves an overall needs-based use of the existing sensor equipment for environmental perception, since computing capacity can be saved by foregoing evaluation of currently less relevant environmental areas.