Autonomous Vehicle Computing Method Prioritization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in efficiently utilizing limited computing resources to run various prediction algorithms for road agent behavior, as different algorithms consume different amounts of resources and have varying levels of accuracy, necessitating a method to prioritize computing methods based on risk and resource consumption.
Innovation Solution
A system and method that receive sensor data, evaluate risk levels, estimate computing resource usage for candidate methods, and select and schedule computing methods to optimize resource allocation while ensuring minimum performance or accuracy, allowing for sequential execution of resource-intensive methods.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the most accurate prediction algorithms are utilized, then the prediction accuracy of road agent behavior is improved, but the computing resource consumption increases
Solution Approach 1:
The system dynamically changes the parameters of computing methods by selecting different prediction algorithms based on risk levels. High-risk inputs are processed with computationally intensive algorithms that provide higher accuracy, while low-risk inputs use simpler algorithms that consume fewer resources. This parameter adjustment resolves the contradiction by adapting the computing method to the specific risk context.
Solution Approach 2:
The system implements dynamic selection of computing methods based on real-time risk assessment. The choice of prediction algorithm is not static but changes dynamically according to the evaluated risk level of each input, allowing the system to optimize between accuracy and resource consumption adaptively.
2Reliability
If multiple prediction algorithms are run simultaneously to ensure accuracy, then the reliability of behavior prediction is improved, but the computing resource consumption increases
Solution Approach 1:
The system applies different qualities of computing methods to different local contexts (risk levels). Instead of uniformly applying multiple algorithms to all inputs, it selectively applies computationally intensive algorithms only where needed (high-risk inputs), while using simpler methods for low-risk inputs. This local differentiation maintains reliability where necessary while conserving resources elsewhere.
Solution Approach 2:
The system performs partial action by selecting only the necessary number of prediction algorithms based on risk levels. Rather than always running all available algorithms, it applies just enough computational effort to achieve reliable predictions for each specific input, avoiding excessive resource consumption.
3Measurement precision
If computationally intensive prediction algorithms are used for all inputs, then the prediction accuracy is improved, but the fuel efficiency decreases
Solution Approach 1:
The system changes the computational parameters based on input risk characteristics. By adjusting the complexity of prediction algorithms dynamically, it achieves high accuracy only when necessary, thereby reducing overall energy consumption and improving fuel efficiency without sacrificing required prediction quality.
Data Source
AI summary
A method includes receiving sensor data associated with one or more inputs associated with a road portion, determining a level of risk associated with each of the one or more inputs, determining an estimated amount of computing resources that each of a plurality of candidate computing methods will consume, and selecting one or more computing methods from the plurality of candidate computing methods to associate with the one or more inputs based on the levels of risk associated with the one or more inputs and the estimated amount of computing resources that the candidate computing methods will consume.