Autonomous Vehicle Motion Control via Object Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in efficiently and safely navigating around proximate objects, as existing systems struggle to accurately identify and respond to potential collision risks and optimize motion plans in real-time, leading to potential safety hazards and inefficiencies.
Innovation Solution
An autonomous vehicle system that processes sensor data to predict the paths of nearby objects, identifies objects of interest based on collision risk, and generates cost data to determine optimal motion plans that adjust speed and trajectory to maintain safe distances, thereby improving navigation and safety.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the autonomous vehicle system processes sensor data to predict paths and identify objects of interest, then the collision avoidance capability is improved, but the computational complexity and processing time increase
Solution Approach 1:
The system segments the processing of proximate objects by identifying specific objects of interest based on collision risk assessment. Rather than processing all detected objects uniformly, the system divides attention into relevant objects (those posing collision risks) and less relevant objects, thereby reducing computational complexity while maintaining collision avoidance reliability.
Solution Approach 2:
The system applies different processing quality levels to different objects based on their relevance. Objects of interest receive detailed path prediction and cost data generation, while other objects receive minimal or no processing. This local differentiation of processing quality reduces overall computational complexity while preserving safety-critical functionality.
2Measurement precision
If the system generates cost data for multiple proximate objects, then the motion planning accuracy is improved, but the real-time response capability deteriorates
Solution Approach 1:
The system segments cost data generation by focusing computational resources only on objects of interest rather than all proximate objects. This selective approach maintains motion planning accuracy for critical objects while reducing overall processing time to meet real-time response requirements.
Solution Approach 2:
The system performs partial cost data generation by calculating costs only for objects that pose collision risks, rather than computing costs for all detected objects. This partial action approach preserves sufficient planning accuracy for safety-critical decisions while reducing computational burden to enable real-time operation.
3Reliability
If the autonomous vehicle continuously adjusts motion based on object proximity, then the safety is improved, but the travel efficiency decreases
Solution Approach 1:
The system applies motion adjustments locally and selectively based on the presence and characteristics of objects of interest. Rather than continuously adjusting motion for all proximate objects, the system modulates speed and trajectory only when necessary to maintain safe distances from objects posing collision risks, thereby preserving travel efficiency while ensuring safety.
Solution Approach 2:
The system performs partial motion adjustments by maintaining safe distances only from objects of interest rather than from all proximate objects. This selective approach ensures safety for critical objects while allowing more aggressive and efficient travel behavior toward less relevant objects, optimizing the trade-off between safety and travel efficiency.
Data Source
AI summary
Systems and methods for controlling the motion of an autonomous are provided. In one example embodiment, a computer implemented method includes obtaining, by one or more computing devices on-board an autonomous vehicle, data associated with one or more objects that are proximate to the autonomous vehicle. The data includes a predicted path of each respective object. The method includes identifying at least one object as an object of interest based at least in part on the data associated with the object of interest. The method includes generating cost data associated with the object of interest. The method includes determining a motion plan for the autonomous vehicle based at least in part on the cost data associated with the object of interest. The method includes providing data indicative of the motion plan to one or more vehicle control systems to implement the motion plan for the autonomous vehicle.