Surrounding Vehicle Trajectory Prediction Using Pre-stored Characteristics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Vehicles face challenges in predicting the trajectory of surrounding vehicles due to sensor failures, which can lead to unsafe maneuvers and potential accidents.
Innovation Solution
A system and method that utilize a sensor system to detect and classify surrounding vehicles, retrieve predetermined vehicle characteristics from a database based on the type, and predict trajectories using both current and predetermined driving characteristics, allowing for safe trajectory determination and fail-safe maneuvers in case of sensor failure.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sensor system is used to monitor external environment, then detection capability is improved, but reliability deteriorates when sensor failure occurs
Solution Approach 1:
The system performs preliminary classification of detected objects into vehicle types (car, truck, motorcycle, etc.) and stores predetermined driving characteristics for each type in advance. When sensor failure occurs, this pre-prepared information allows the system to continue predicting trajectories of surrounding vehicles using the stored characteristics rather than real-time sensor data, thereby maintaining operational reliability despite sensor failure.
Solution Approach 2:
The patent introduces a database as an intermediary component that stores predetermined driving characteristics for different vehicle types. This database acts as a mediator between the sensor system and the trajectory prediction system, allowing the prediction to continue using stored characteristic data when direct sensor input becomes unreliable or unavailable due to failure.
2Reliability
If fail-safe maneuvers are executed upon sensor failure, then safety is improved, but productivity deteriorates due to restricted operation
Solution Approach 1:
Instead of completely restricting operation upon sensor failure, the system applies partial action by using predetermined driving characteristics for specific vehicle types to make reasonable trajectory predictions. This allows the vehicle to continue operating with reduced but sufficient functionality - enough to ensure safety while maintaining the ability to navigate and reach destinations, rather than forcing a complete stop or fail-safe maneuver.
3Adaptability or versatility
If trajectory prediction uses only current driving characteristics, then adaptability is improved, but reliability deteriorates when sensors fail
Solution Approach 1:
The system merges two sources of information: current driving characteristics obtained from real-time sensor data and predetermined driving characteristics stored in the database for the identified vehicle type. By combining these two sources, the system maintains adaptability to current behavior while ensuring reliability through the backup of predetermined characteristics, especially when sensor data becomes unavailable due to failure.
Data Source
AI summary
A host vehicle can be configured to monitor an external environment of the host vehicle using a sensor system. The host vehicle can detect an object in the external environment of the host vehicle. The object can be classified as a surrounding vehicle. The host vehicle can determine a driving behavior of the surrounding vehicle. The driving behavior can be based on any one of a predetermined vehicle characteristics and current driving characteristics of the surrounding vehicle. The predetermined vehicle characteristics can be based on a type of the surrounding vehicle. The current driving characteristics can be based on observations obtained by the sensor system. The host vehicle can predict a trajectory of the surrounding vehicle based on the determined driving behavior. The host vehicle can follow a safe path based on the predicted trajectory responsive to detecting one or more sensors in a failed state.