In-Vehicle Cause Feedback for Adaptive Autonomous Driving Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current autonomous vehicle systems lack the ability to effectively manage situations where computer-based control decisions are unnecessary, leading to increased driver burden and potential safety issues, as they cannot grasp human intentions or context, such as pedestrians conversing at a crossing.
Innovation Solution
An in-vehicle device that monitors external objects and states using sensors, outputs cause information for driver awareness, and allows input changes to control processes, enabling continued autonomous driving without manual intervention by notifying and allowing cancellation or modification of control actions based on driver inputs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If autonomous travel control is implemented with computer-based decision making, then automation extent is improved, but driver burden increases when manual switching is required
Solution Approach 1:
The system provides feedback to the driver by outputting the cause information through the output unit, allowing the driver to understand why autonomous control decisions are made. This feedback mechanism enables the driver to monitor the system's reasoning without needing to constantly switch to manual control, thereby maintaining high automation while reducing driver burden.
Solution Approach 2:
The autonomous travel control device independently determines driving behavior and its causes without requiring driver intervention. The system serves itself by autonomously making decisions based on sensor data and pre-stored rules, only seeking driver input when the determination unit identifies that manual input is needed, thus maintaining automation while minimizing burden.
2Ease of operation
If autonomous control continues without manual switching, then ease of operation is improved, but reliability decreases due to inability to grasp human intentions
Solution Approach 1:
The system dynamically adjusts the level of autonomy based on the situation. The control unit switches between fully autonomous operation and manual control modes depending on whether the determination unit identifies that driver input is needed. This dynamic adaptation allows the system to maintain ease of operation in routine situations while ensuring reliability when human judgment is required.
Solution Approach 2:
The determination unit acts as an intermediary between the autonomous control system and the driver. It evaluates whether the current situation requires human intervention and mediates between continuing autonomous control and switching to manual control, thereby balancing ease of operation with reliability by involving the driver only when necessary.
3Reliability
If driver input is required for every situation, then reliability is improved, but productivity decreases due to frequent manual intervention
Solution Approach 1:
Instead of requiring driver input for every situation (excessive action), the system applies driver input only partially - specifically when the determination unit identifies situations where human judgment may be needed. This partial application of manual control maintains reliability while preserving autonomous driving efficiency in the majority of routine situations.
4Measurement precision
If the system outputs detailed cause information, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The system segments the cause identification function into distinct components: the sensor unit collects data, the determination unit analyzes whether driver input is needed, and the output unit presents cause information to the driver. This segmentation allows detailed cause identification while managing complexity by distributing functions across specialized units with clear interfaces.
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
The present invention provides an in-vehicle device (10) that includes a monitoring unit (11) that determines a state of an object on the basis of an output of a sensor mounted in a vehicle, a generation unit (12) that generates process information for causing the vehicle to execute a process in accordance with the state, a control unit (13) that causes an output device to output cause information indicating at least one of the object or the state which is a cause of the process, and a reception unit (14) that receives an input for changing the process due to the cause indicated by cause information, in which, on the basis of a reception result of the reception unit (14), the generation unit (12) generates the process information in which the process is changed.