Vehicle Sensor Fusion for Concealed Object Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Radar sensors face challenges in detecting dynamic objects behind obstacles, such as static objects or weather-related effects, which can lead to increased collision probability due to incomplete scanning ranges, resulting in potential collisions.
Innovation Solution
A method that uses radar and LIDAR sensors, combined with optical sensors and digital maps, to detect obstacles and predict potential dynamic objects in concealed sections of the scanning range, adapting vehicle behavior to mitigate collision risks by generating a probability-based object prediction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If radar sensors are used to detect objects in the scanning range, then object detection capability is improved, but detection reliability deteriorates when obstacles block the scanning range
Solution Approach 1:
The system performs preliminary detection of obstacles before they completely block the scanning range. By identifying obstacles early and predicting their impact on the scanning range, the system can prepare alternative detection strategies and adapt driving behavior in advance, maintaining reliability even when detection is partially blocked
Solution Approach 2:
The control unit acts as an intermediary that processes radar data, identifies blocked scanning ranges, and generates predictions about concealed objects. This intermediary processing layer transforms raw radar measurements into actionable safety decisions, bridging the gap between partial detection and reliable collision avoidance
2Loss of information
If the scanning range is blocked by obstacles, then information about dynamic objects is lost, but the system can adapt driving behavior to compensate
Solution Approach 1:
The system performs preliminary identification of blocked scanning ranges and generates predictions about potential concealed objects before actual collision risks arise. This advance preparation allows the system to maintain collision avoidance reliability despite information loss by having prediction models ready and driving behavior already adapted
Solution Approach 2:
The system converts the harmful effect of blocked scanning ranges into a benefit by using obstacle detection to trigger prediction algorithms. The obstruction itself becomes a signal that activates enhanced prediction modes, transforming information loss into an opportunity for more conservative and safe driving behavior adaptation
3Area of stationary object
If radar beams are blocked by weather effects or static objects, then the scanning range is reduced, but the system can use multiple sensor types to compensate
Solution Approach 1:
The system merges radar sensor data with LIDAR and optical sensor data to create a comprehensive view of the environment. By combining multiple sensor types, the system compensates for reduced effective scanning range caused by blockages, maintaining detection reliability through multi-sensor fusion and cross-validation of detections
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enhances traffic safety by enabling situation-adaptive vehicle behavior, particularly in highly automated and fully automated driving scenarios, by accounting for concealed areas and potential object trajectories, thereby reducing collision probabilities.
Implementation Method 1
Radar sensors and the possibility of using the Doppler effect make it possible to perform relative radial speed measurements between a radar sensor and a reflecting object
Implementation Method 2
measurement data collected by a radar sensor and/or LIDAR sensor
Data Source
AI summary
A method for the adaptation of a driving behavior of a vehicle by a control unit. Measurement data collected by at least one radar sensor and/or LIDAR sensor are received and, by evaluating the received measurement data, at least one obstacle within a scanning range of the radar sensor and/or LIDAR sensor is ascertained. Based on the ascertained obstacle and an installation position of the radar sensor and/or LIDAR sensor, a section of the scanning range concealed by the obstacle is ascertained. An object prediction is generated for the concealed section of the scanning range that a dynamic object may potentially cross a driving lane of the vehicle from the concealed section of the scanning range. A driving behavior of the vehicle is adapted as a function of the situation based on the object prediction. A control unit, a computer program, and a machine-readable storage medium are also described.