Autonomous Vehicle Acceleration Control During Leading Vehicle Lane Changes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous driving systems face challenges in smoothly controlling the longitudinal acceleration of a host vehicle relative to a road, especially when interacting with leading vehicles during lane changes, which can result in uncomfortable driving experiences like sharp jerks.
Innovation Solution
A computer-implemented method and apparatus that determine a target longitudinal acceleration for a host vehicle by defining a lateral range extending from a leading vehicle, with different acceleration values set based on the host vehicle's lateral position within this range, ensuring smooth acceleration control during lane changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If the autonomous driving system uses a fixed acceleration control strategy, then the control logic is simple, but the driving comfort deteriorates due to sharp jerks during lane changes
Solution Approach 1:
The patent applies local quality by defining different acceleration values based on the host vehicle's lateral position within the lateral range. The lateral range is divided into subranges (first lateral subrange, second lateral subrange, and central lateral subrange), and different acceleration values are assigned to different subranges. This allows the acceleration control to be locally optimized for comfort while maintaining overall system simplicity.
2Device complexity
If the lateral range is fixed, then the control system is simple, but the ability to handle lane change scenarios deteriorates
Solution Approach 1:
The patent implements dynamics by making the lateral range variable rather than fixed. The lateral range extends from the lateral position of the leading vehicle in the direction of lane change, and its extent is dynamically adjusted based on the lane change scenario. This dynamic adjustment allows the system to adapt to different lane change situations while maintaining a relatively simple control structure.
3Device complexity
If the acceleration value is uniformly set, then the control implementation is simple, but the smoothness of acceleration control deteriorates
Solution Approach 1:
The patent applies parameter changes by setting different acceleration values based on the host vehicle's lateral position within different subranges of the lateral range. Specifically, a first acceleration value is set for the first lateral subrange, a second acceleration value is set for the second lateral subrange, and a third acceleration value is set for the central lateral subrange. This parameter variation ensures smooth acceleration control while keeping the control implementation relatively simple.
Data Source
AI summary
A method of setting a target longitudinal acceleration of a vehicle travelling along a road behind a leading vehicle, comprising: determining a lateral position of the leading vehicle; defining a lateral range extending from the leading vehicle and having a first subrange, a second subrange and a central subrange therebetween, wherein the lateral range increases with lateral distance of the leading vehicle from a centre of a lane in which the leading vehicle is located when the leading vehicle is changing lanes; determining a longitudinal range extending behind the leading vehicle; and setting the target longitudinal acceleration, for any longitudinal position of the host vehicle within the longitudinal range, to: a first value if a lateral position of the host vehicle is within the central subrange; and a second value greater than the first value if the lateral position is within the first or second subrange.