Autonomous Vehicle Acceleration Limiting for Nearby Lateral Objects
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle acceleration control systems do not effectively incorporate lateral object information, leading to uncomfortable acceleration when passing nearby moving objects, which can result in a poor ride quality.
Innovation Solution
A method and system that detect lateral distances to objects using sensor data, compare these distances to a threshold value, and selectively limit acceleration based on the comparison, ensuring comfortable passage of nearby objects by adjusting acceleration limits accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If pre-defined maximum acceleration values or road condition-based limits are used, then acceleration control is simplified, but the vehicle accelerates uncomfortably when passing moving objects nearby
Solution Approach 1:
The system dynamically adjusts acceleration limits based on real-time lateral object detection and classification. Instead of using fixed pre-defined maximum values, the acceleration control adapts to current driving conditions by detecting objects in the lateral direction, classifying them (pedestrian, cyclist, vehicle, animal, or other), and selectively applying acceleration limits only when appropriate objects are detected nearby, thereby maintaining ride comfort without excessive complexity
Solution Approach 2:
The system introduces an intermediary object classification mechanism between the sensor data and acceleration control. The lateral object detector and classifier act as intermediaries that process sensor data to identify and categorize objects, then use this classification information to modulate acceleration limits. This intermediary layer enables nuanced control decisions based on object type, resolving the contradiction between simple control and comfortable operation
2Reliability
If acceleration limiting is applied continuously, then safety around pedestrians is improved, but the vehicle acceleration becomes overly restricted even when no lateral objects are present
Solution Approach 1:
The system applies acceleration limiting locally and selectively based on the presence and classification of lateral objects. Instead of continuous acceleration limiting, the control is applied only in specific situations where classified objects (pedestrians, cyclists, vehicles, animals) are detected within a threshold lateral distance. This local application of quality control ensures safety around vulnerable road users while maintaining normal acceleration performance when no such objects are present
Solution Approach 2:
The system changes the acceleration parameter dynamically based on detected object classification. When a relevant object is detected and classified, the acceleration limit parameter is adjusted to a more restrictive value; when no relevant objects are present, the parameter returns to its normal unrestricted value. This parameter change approach ensures safety when needed while preserving speed and acceleration performance when safe to do so
Data Source
AI summary
Disclosed herein are systems, methods, and computer program products for controlling acceleration of a vehicle. The methods comprising: detecting a lateral distance from a point on a trajectory of the vehicle to a first object the vehicle is expected to pass when following the trajectory; selecting whether acceleration limiting is to be performed by the vehicle based on the lateral distance; obtaining a margin of the vehicle defined by a sequence of points; obtaining an amount by which the acceleration of the vehicle is to be limited based on the trajectory and the margin of the vehicle, when a selection is made that acceleration limiting is to be performed by the vehicle; and causing the vehicle to perform operations for autonomous driving with limiting of acceleration by the obtained amount.