Autonomous Vehicle Position Correction for Smoothness
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional drive-assisted vehicle systems prioritize 'non-departing/non-colliding' over 'smoothness' when correcting vehicle position, leading to unnecessary large corrections and compromised smoothness.
Innovation Solution
A method and device that corrects the target route of a drive-assisted vehicle by allowing sideways movement based on detected lane boundaries and vehicle speed, enabling selection between prioritizing smoothness or non-departure depending on the scenario, using onboard sensors and navigation control units to adjust the target route laterally.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the self-position of the host vehicle is corrected to prioritize non-departing/non-colliding, then safety is improved, but smoothness of vehicle behavior deteriorates due to unnecessary large corrections
Solution Approach 1:
The patent segments the position correction function into two independent parts: (1) self-position correction that maintains safety by ensuring the vehicle does not depart from or collide with lane boundaries, and (2) target route correction that optimizes smoothness by adjusting the reference trajectory. This segmentation allows each component to be optimized independently for its specific objective.
Solution Approach 2:
The patent introduces a target route as an intermediary element between the self-position and the final vehicle behavior. Instead of directly correcting the self-position which affects all downstream modules, the correction is applied to the target route which serves as a reference for path following controllers. This intermediary approach isolates the safety-critical position estimation from the behavior-critical trajectory generation.
2Reliability
If large corrections are applied to ensure non-departure from lane, then lane boundary compliance is improved, but vehicle behavior becomes unnatural and uncomfortable
Solution Approach 1:
The patent applies different correction qualities to different components: the self-position correction uses conservative, safety-oriented logic with smaller correction amounts to maintain stability, while the target route correction uses aggressive, behavior-optimized logic to ensure lane boundary compliance. Each component has its own correction characteristics suited to its specific function.
Solution Approach 2:
The patent implements dynamic correction strategies where the self-position correction amount is limited to prevent excessive adjustments that would cause unnatural behavior, while the target route correction dynamically adjusts the reference trajectory to maintain optimal lane positioning. The system adapts correction magnitudes based on current driving conditions and error magnitudes.
3Measurement precision
If the target route is corrected with unrestricted lateral movement speed, then position accuracy is improved, but vehicle responsiveness and comfort deteriorate
Solution Approach 1:
The patent implements periodic or incremental correction adjustments rather than instantaneous full-correction. The target route is adjusted in controlled steps with rate limiting on lateral movement speed, creating a smooth transition that maintains position accuracy while respecting vehicle dynamics and driver comfort. This staged correction approach prevents abrupt changes that would compromise responsiveness.
Solution Approach 2:
The patent changes the parameter of lateral movement speed from an unrestricted value to a rate-limited value with maximum bounds. By constraining the rate of change of the target route's lateral position, the system ensures that corrections improve position accuracy without causing abrupt vehicle responses that would reduce comfort and perceived responsiveness.
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
The selection of whether to prioritize smoothness or to prioritize non-departure, depending on the scenario, is made possible, and vehicle behavior where it is possible to feel more at ease is achieved. In a device for correcting a position error of an autonomous vehicle provided with a navigation control unit (3) that corrects error during autonomous travel, the navigation control unit (3) includes, in a target route corrector 36 that corrects a target route, a road boundary information consolidation unit (361), a lateral correction amount calculation unit (362), and a lateral/sideways movement unit (363). The road boundary information consolidation unit (361) detects a lane boundary of a lane in which the host vehicle travels. The lateral correction amount calculation unit (362) compares positional relationships between a lane boundary that was detected and a target route on a map, and in situations where the target route is within a prescribed distance of the lane boundary, or in situations where the target route is on the opposite side of the lane boundary to the host vehicle, calculates a lateral correction amount for the target route. The lateral/sideways movement unit (363), upon the calculation of the lateral correction amount, moves the target route sideways in a lateral direction by the lateral correction amount to correct the target route.