Autonomous Driving Algorithm Updates From Manual Driving Paths
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous driving systems fail to adapt to individual driver habits, compromising the driving experience and reliability.
Innovation Solution
An autonomous driving system updates its planning and control algorithms by comparing the driver's manual driving path with sensor data to adjust its behavior, using a first sensor group for environment sensing and a second sensor group for vehicle dynamics, with cloud-based evaluation for optimal adaptation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the autonomous driving system is updated to adapt to individual driver habits, then the driving experience is improved, but the reliability of planning and decision-making may be compromised
Solution Approach 1:
The system performs preliminary learning of driver habits during manual driving mode before applying adaptations in autonomous mode. Sensor data is collected and analyzed in advance to establish driver behavior patterns, ensuring that adaptations are based on verified habitual behaviors rather than random variations, thus maintaining reliability while improving adaptability.
Solution Approach 2:
The system implements a feedback mechanism where the autonomous driving system continuously monitors and compares its planned actions with the driver's actual manual driving actions. This feedback loop allows the system to learn and adapt to driver habits while maintaining safety through continuous verification against established behavioral patterns, resolving the contradiction between adaptability and reliability.
2Adaptability or versatility
If the autonomous driving system collects and processes extensive sensor data for updating, then the adaptability to driver habits is improved, but the device complexity increases
Solution Approach 1:
The system extracts only the essential and relevant features from extensive sensor data for updating the autonomous driving system. Instead of processing all raw sensor data, the system identifies and extracts key behavioral patterns and parameters that characterize driver habits, significantly reducing processing complexity while maintaining adaptability to individual driving styles.
Solution Approach 2:
The data processing system is segmented into multiple specialized modules, each handling specific types of sensor data or specific aspects of driver behavior analysis. This segmentation allows parallel processing of different data streams (e.g., steering patterns, braking patterns, acceleration patterns) independently, reducing overall system complexity while comprehensively capturing driver habits.
Data Source
Figure 1~2
Figure 3~4
AI summary
Embodiments of the present disclosure relate to the technical field of autonomous driving, and in particular to methods for updating an autonomous driving system, autonomous driving systems, and on-board apparatuses. In the embodiments of the present disclosure, the autonomous driving system, even when in a manual driving mode, also senses the surrounding environment of a vehicle, performs vehicle positioning, and plans a path for autonomous driving for the vehicle according to environment sensing information, positioning information, and data of vehicle sensors. However, the autonomous driving system does not issue an instruction to control the driving of the vehicle. Instead, it compares the path with a path along which a driver drives the vehicle in the manual driving mode to update a planning and control algorithm of the autonomous driving system. As such, the updated autonomous driving system better caters to the driving habits of the driver and improves the driving experience for the driver without compromising the reliability of planning and decision-making of autonomous driving.