Autonomous Driving Risk Maps for Occupant Comfort Around Obstacles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing driving assistance systems for autonomous vehicles do not adequately consider the feelings and comfort of all occupants, leading to anxiety and discomfort during obstacle avoidance maneuvers, as they primarily focus on the driver's preferences and do not account for the sensations of other passengers.
Innovation Solution
A driving assistance device that learns the driving characteristics of individual drivers during manual driving and sets vehicle risk maps for autonomous driving, incorporating both driver and obstacle risk potentials to optimize vehicle trajectories and speed, thereby reducing anxiety and discomfort for all occupants by minimizing collision risks and adjusting to their perceived danger levels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If the driving assistance system focuses on driver's preferences and uses standard risk assessment models, then the system complexity remains manageable, but the comfort and anxiety reduction for all occupants (especially non-driver passengers) is insufficient
Solution Approach 1:
The risk assessment is segmented into two distinct components: driver risk preference (learned from driver's manual driving behavior) and passenger risk sensitivity (detected through passenger monitoring). This segmentation allows the system to handle different occupant needs separately, improving overall comfort without requiring a complete system redesign.
Solution Approach 2:
The system performs preliminary learning of the driver's risk preferences during manual driving phases before autonomous driving begins. This preliminary action stores the driver's characteristics in advance, so that when autonomous driving starts, the system can immediately apply personalized risk assessment without real-time computation delays, reducing complexity during critical driving moments.
2Reliability
If the system uses a single risk threshold for all occupants, then the control logic remains simple, but it cannot account for individual perceptions of danger and causes anxiety for risk-sensitive passengers
Solution Approach 1:
The system applies local quality by setting different risk thresholds for different occupants based on their individual characteristics. The driver's risk preference and passenger's risk sensitivity are used to create occupant-specific risk thresholds, allowing each occupant to experience driving behavior matched to their personal comfort level rather than a uniform approach.
Solution Approach 2:
The system continuously monitors passenger state and adjusts driving behavior feedback based on detected passenger risk sensitivity. When a passenger is identified as risk-sensitive, the system provides feedback by modifying autonomous driving actions to be more conservative, creating a closed-loop system that adapts to passenger needs in real-time.
3Ease of operation
If the autonomous driving system adopts conservative driving behavior to reduce passenger anxiety, then passenger comfort improves, but the overall driving efficiency and productivity decrease
Solution Approach 1:
The system dynamically adjusts driving behavior based on real-time assessment of passenger risk sensitivity rather than maintaining a fixed conservative approach. When passengers are detected to be risk-tolerant or not present, the system can adopt more efficient driving patterns. This dynamic adaptation allows the system to optimize between comfort and efficiency on a case-by-case basis.
Solution Approach 2:
The system changes key driving parameters (such as deceleration rate, lateral acceleration, and following distance) based on the detected passenger risk sensitivity. Instead of always using conservative parameter values, the system adjusts these parameters dynamically according to the specific passenger's tolerance level, thereby maintaining efficiency when possible while ensuring comfort when needed.
Data Source
AI summary
A driving assistance device reflects, in autonomous driving control on a vehicle, a driving characteristic of a driver who drives the vehicle learned during manual driving of the vehicle. The driving assistance device includes a storage that stores information on an individual risk potential set for the vehicle by learning a risk sensed by the driver for each of one or more obstacles during the manual driving, a vehicle risk calculator that sets, for the vehicle, a vehicle risk map that reflects the individual risk potential of an occupant of the vehicle during autonomous driving of the vehicle, and a driving condition setter that sets a driving condition for the autonomous driving of the vehicle based on information on the vehicle risk map and information on an obstacle risk map that reflects an obstacle risk potential set for the each of the one or more obstacles.