Vehicle Intention Prediction Display for Real-Time Collision Avoidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current safety features in autonomous vehicles may not adequately enhance safety due to information overload and the inability to effectively predict and share vehicle moving intentions in real-time, leading to potential collisions.
Innovation Solution
Implementing a computer-implemented method for predicting vehicle moving intentions, broadcasting and sharing these predictions through a predefined network, and mapping them onto a 2D or 3D spatial map for real-time display to vehicle operators using augmented reality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If current safety features are implemented in autonomous vehicles, then basic safety monitoring is provided, but information overload occurs and the ability to effectively predict and share vehicle moving intentions in real-time deteriorates
Solution Approach 1:
The patent extracts and separates critical moving intention prediction information from the overall data stream. By using dedicated V2X communication channels to broadcast specific intention data (steering angle, acceleration, braking) separately from general sensor data, the system prevents information overload while maintaining comprehensive safety monitoring capabilities.
Solution Approach 2:
The system performs preliminary prediction of vehicle moving intentions using machine learning models before the actual movement occurs. By analyzing current sensor data (steering wheel position, accelerator pedal position, brake pedal position) and predicting future intentions, the system prepares safety responses in advance, reducing the processing burden during critical moments.
2Measurement precision
If comprehensive sensor data is collected for accurate prediction, then prediction accuracy improves, but real-time processing and sharing capability deteriorates
Solution Approach 1:
The patent segments the comprehensive sensor data into distinct categories (steering input, acceleration input, braking input, vehicle state) and processes each segment through specialized machine learning models. This segmentation allows parallel processing of multiple data streams simultaneously, maintaining prediction accuracy while achieving real-time processing speeds through distributed computation.
Solution Approach 2:
The system introduces V2X communication as an intermediary layer between data collection and prediction processing. By broadcasting predicted intentions through standardized communication protocols, the system enables other vehicles to receive and process the information without requiring them to handle the full raw sensor data stream, thus maintaining real-time performance.
3Object-affected harmful factors
If vehicle moving intentions are predicted and shared with nearby vehicles, then collision risk reduction improves, but network communication overhead and complexity increases
Solution Approach 1:
The patent transforms complex multi-dimensional sensor data into simplified parametric representations of vehicle intentions (predicted steering angle, predicted acceleration, predicted braking force, time-to-collision estimates). By changing the data parameters from raw sensor readings to standardized intention metrics, the system reduces communication overhead while maintaining the ability to assess and respond to collision risks effectively.
Data Source
AI summary
In an approach to improve manual and autonomous operations of a vehicle embodiments predict a moving intention of a first vehicle. Based on the predicted moving intention embodiments broadcast and share the predicted vehicle moving intention of the first with a set of key vector data through a predefined network to a second vehicle. Further, embodiments receive the broadcasting signal embedded with the set of key vector data related to vehicle moving intention nearby and map the received vehicle moving intentions to two-dimensional (2d) or three-dimensional (3D) spatial map. Additionally, embodiments render the 2D or 3D spatial map with the mapped vehicle moving intentions in a virtual display.