Comfort Scale Modeling for Pedestrian-Aware Autonomous Driving
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in fully autonomous driving due to the complexity of interacting with pedestrians and other road users, as existing metrics like speed and proximity are inadequate and disruptive, and unwritten traffic rules vary significantly by location, making it difficult to achieve smooth and courteous navigation.
Innovation Solution
A system that automatically generates comfort scales representing subjective experiences of road users by using trip log data and multiple rater pools to model and transform ratings, allowing for training of driving control models to account for localized and temporal variations in unwritten driving norms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If autonomous vehicles use simple metrics like speed and proximity to make driving decisions, then the control system is simple to implement, but the driving behavior becomes disruptive and inefficient
Solution Approach 1:
The patent transforms the control system from using simple physical parameters (speed, proximity) to using subjective experience parameters (comfort scales). By changing the parameter space from objective measurements to modeled subjective experiences, the system achieves both simplicity and effectiveness in driving decisions.
2Reliability
If autonomous vehicles stop or slow down whenever a pedestrian is within threshold distance, then pedestrian safety is improved, but traffic flow efficiency deteriorates
Solution Approach 1:
The system uses comfort scale ratings as feedback to determine appropriate vehicle responses. Instead of fixed threshold-based reactions, the control system continuously monitors modeled pedestrian comfort levels and adjusts driving behavior accordingly, allowing safe operation while maintaining traffic flow.
3Reliability
If autonomous vehicles adhere strictly to written traffic rules, then legal compliance is ensured, but social norms and unwritten rules cannot be followed
Solution Approach 1:
The system dynamically adjusts driving behavior based on contextual factors including location, time of day, and day of week. By making the control parameters dynamic rather than static, the vehicle can adapt to local unwritten norms while maintaining compliance with written rules, achieving both legal and social appropriateness.
4Stability of the object's composition
If autonomous vehicles use fixed driving norms, then consistent behavior is achieved, but localized and temporal variations in traffic expectations cannot be accommodated
Solution Approach 1:
The patent implements location-specific and time-specific comfort scale models that capture local and temporal variations in pedestrian expectations. Different geographic locations and time periods have their own calibrated comfort parameters, allowing the vehicle to adapt behavior to local conditions while maintaining overall system consistency.
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on computer storage media, for using comfort scales to assess the performance of autonomous vehicles. One of the methods includes receiving data representing a traffic encounter between a vehicle and a pedestrian. A plurality of comfort scale ratings of the encounter assigned by a rater belonging to a first rater pool are received. An input element is generated for a rating transformation model configured to predict how comfort scale ratings assigned by a particular rater pool would have been assigned by a representative rater belonging to a reference rater pool. An inference pass is performed over the rating transformation model using the input element to obtain a plurality of transformed comfort scale ratings for the reference rater pool.