Autonomous Driving Risk Assessment via Trajectory Error Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in maintaining driving stability and accuracy due to the lack of effective risk assessment and passenger involvement in driving control, leading to potential collisions and unreliable autonomous driving control.
Innovation Solution
An autonomous driving apparatus and method that includes a sensor unit, memory, and processor to detect surrounding vehicles, generate driving trajectories, assess autonomous driving risks, and output warnings to passengers, while also learning from passenger driving manipulation to improve algorithm accuracy and control vehicle states based on internal modes set by passengers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If autonomous driving control is implemented without effective risk assessment, then automation level is improved, but driving stability and safety deteriorate
Solution Approach 1:
The system continuously monitors surrounding vehicles' driving trajectories and compares actual trajectories with expected trajectories to assess autonomous driving risks. This feedback mechanism enables real-time risk assessment and warning level adjustment, improving driving stability while maintaining automation.
Solution Approach 2:
The system performs preliminary risk assessment by analyzing surrounding vehicle trajectories before critical situations occur. By generating warnings at three levels based on trajectory errors, the system prepares for potential collisions in advance, enhancing reliability without reducing automation extent.
2Extent of automation
If passenger is not involved in driving control, then automation level is improved, but driving accuracy deteriorates
Solution Approach 1:
The system incorporates passenger feedback through warning outputs that communicate assessed risks to the passenger. This allows the passenger to provide implicit feedback through their responses, improving driving accuracy while maintaining high automation levels by only engaging the passenger when necessary.
Solution Approach 2:
The system serves itself by automatically assessing risks and determining appropriate warning levels without requiring continuous passenger intervention. The autonomous system monitors trajectory errors and self-adjusts warning outputs, maintaining automation while improving accuracy through intelligent risk assessment.
3Reliability
If warning system is implemented with multiple levels, then driving stability is improved, but device complexity increases
Solution Approach 1:
The system uses parameter changes in trajectory error values to determine warning levels. By monitoring changes in actual versus expected trajectory parameters, the system implements a three-level warning structure that improves reliability without requiring complex additional hardware, only sophisticated parameter analysis.
4Measurement precision
If autonomous driving risk assessment is performed continuously, then driving accuracy is improved, but energy consumption increases
Solution Approach 1:
The system performs risk assessment periodically by monitoring trajectory errors at regular intervals rather than continuously. This periodic action maintains driving accuracy through sufficient sampling while reducing energy consumption by allowing processing cycles to occur at optimized intervals rather than constant real-time monitoring.
Data Source
AI summary
An autonomous driving apparatus and method, in which the apparatus includes a sensor unit detecting a surrounding vehicle around an ego vehicle that autonomously travels, an output unit, a memory storing map information, and a processor controlling the autonomous driving of the ego vehicle based on the map information. The processor is configured to generate an actual driving trajectory and expected driving trajectory of the surrounding vehicle based on driving information of the surrounding vehicle detected by the sensor unit and the map information, determine whether a driving mode of the surrounding vehicle is an autonomous driving mode and an autonomous driving risk of the ego vehicle based on a trajectory error between the actual driving trajectory and expected driving trajectory of the surrounding vehicle, and output a warning to a passenger through the output unit at a level corresponding to the determined autonomous driving risk.