Event-Based Vehicle Hazard Detection for Impending Object Motion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current vehicle safety systems lack the necessary speed and accuracy to prevent collisions, as they often rely on camera-based sensors that require complete image frames to detect hazards, leading to delayed responses in potentially critical situations.
Innovation Solution
The method employs event-based sensors with light-sensitive pixels to detect and analyze on-the-spot movements and changes in the vehicle's surroundings, allowing for quicker identification of hazardous or at-risk objects by outputting events based on predefined intensity changes, which can be combined with camera data for enhanced object classification and tracking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If camera-based sensors are used to detect hazards, then complete image frames can be analyzed for accurate object identification, but the response time is delayed due to the requirement of waiting for complete image frames
Solution Approach 1:
The patent segments the image acquisition process by using event-based sensors that detect and process individual pixel changes independently rather than waiting for complete image frames. This allows the system to process only the relevant portions of visual information that contain hazard-related events, achieving fast response times while maintaining identification accuracy through selective processing of high-priority visual data
Solution Approach 2:
The system dynamically adjusts its processing approach by using event-based sensors that continuously monitor for changes in the visual scene and trigger processing only when hazards are detected. This dynamic event-driven architecture allows the system to maintain high measurement precision for hazard identification while achieving rapid response times by processing visual data in real-time as changes occur rather than in fixed frame intervals
2Loss of time
If event-based sensors are used to detect hazards, then response time is reduced by processing individual pixel events, but the complexity of processing event streams increases
Solution Approach 1:
The patent extracts only the essential hazard-related information from the event stream by focusing processing on specific pixel events that indicate potential hazards. Rather than processing the complete event stream in full detail, the system extracts and prioritizes only those events that are relevant to hazard detection, thereby reducing processing complexity while maintaining fast response times for critical safety functions
Solution Approach 2:
The system applies different processing quality levels to different regions of the visual field by prioritizing processing of events in hazard-relevant areas. Events occurring in regions where hazards are more likely to occur receive more detailed processing, while other areas receive minimal processing. This local quality approach reduces overall processing complexity while maintaining high response times for critical hazard detection
3Loss of information
If complete image frames are processed, then all objects in the scene can be identified, but bandwidth requirements increase due to transmitting and processing large amounts of image data
Solution Approach 1:
The patent extracts and transmits only the essential visual information related to hazards rather than complete image frames. By extracting only the relevant pixel events that indicate potential hazards and transmitting this compressed event stream, the system maintains complete object detection capability for safety-critical functions while dramatically reducing bandwidth consumption compared to transmitting full-resolution image data
Solution Approach 2:
The system processes and transmits only the partial visual information that is necessary for hazard detection rather than complete image frames. This partial action approach focuses computational and communication resources exclusively on detecting hazards and related objects, achieving sufficient object detection completeness for safety purposes while minimizing bandwidth requirements by avoiding transmission of redundant visual data
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
This approach enables faster and more detailed detection of subtle movements, reducing response time and bandwidth requirements, allowing for timely interventions to avoid collisions by assessing impending movements and changes of state, thereby improving vehicle safety.
Implementation Method 1
at least one event-based sensor including light-sensitive pixels, in that a relative change of the light intensity incident upon a pixel by at least a predefined percentage prompts the sensor to output an event
Data Source
AI summary
A method for identifying potentially hazardous or at-risk objects in the surroundings of a vehicle. The method includes detecting an area of the surroundings using at least one event-based sensor, the event-based sensor including light-sensitive pixels, and a relative change of the light intensity incident upon a pixel by at least a predefined percentage prompting the sensor to output an event assigned to this pixel. The method also includes assigning events output by the sensor to objects in the area; analyzing, for at least one object to which events are assigned, the events assigned to the object with respect to present movements of the object; and ascertaining an impending movement of the object, and/or an impending change of state, of the object from the present movements. An associated computer program is also described.