Automatic Parking Gradient Estimation for Precise Longitudinal Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous vehicle parking systems face challenges in controlling vehicle speed and braking due to the gradient of parking spaces, particularly in inclined environments, leading to increased parking control time and inefficiencies.
Innovation Solution
A method and device that estimate the gradient of a parking space using parking line recognition information to adjust driving torque and braking pressure based on the angular relationships between parking lines, enabling precise longitudinal control of the vehicle.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If high driving torque is used to maintain vehicle speed in inclined parking environments, then vehicle speed control is achieved, but high braking pressure is required which increases parking control time
Solution Approach 1:
The system performs preliminary gradient estimation by analyzing parking line angles from captured images before the actual parking maneuver. This advance knowledge of the parking space gradient allows the control system to pre-calculate appropriate driving torque and braking pressure values, eliminating the need for reactive adjustments during parking execution and thereby reducing overall parking control time
Solution Approach 2:
The system continuously monitors the actual parking space gradient during the parking maneuver and compares it with the estimated gradient. Based on this feedback, the control system dynamically adjusts driving torque and braking pressure to maintain optimal vehicle speed, preventing excessive braking pressure application and reducing unnecessary control time delays
2Speed
If gradual braking pressure reduction is applied upon re-starting to control vehicle speed below creep speed, then vehicle speed control precision is improved, but parking control time increases
Solution Approach 1:
The system dynamically adjusts braking pressure based on real-time gradient measurements and vehicle speed feedback. Instead of applying a fixed gradual reduction schedule, the braking pressure is continuously optimized according to the actual gradient conditions and current speed, allowing for faster yet precise speed control that reduces parking time while maintaining control accuracy
Solution Approach 2:
The system changes the braking pressure parameter dynamically based on the estimated gradient and vehicle state. By calculating optimal braking pressure values that account for gradient effects, the system achieves precise speed control below creep speed without requiring extended gradual reduction periods, thereby reducing overall parking control time
3Measurement precision
If parking line recognition information is used to estimate gradient, then gradient estimation accuracy is improved, but system complexity increases
Solution Approach 1:
The system uses the parking line recognition information for multiple purposes: it identifies parking space boundaries for navigation and simultaneously estimates the gradient for torque control. By making the parking line detection module multi-functional, the system avoids adding separate gradient measurement hardware, thus improving gradient estimation accuracy without significantly increasing system complexity
Solution Approach 2:
The system uses its own existing parking line recognition capability to estimate gradient, rather than relying on external or additional sensors. The parking line angle information, already captured for navigation purposes, is repurposed for gradient calculation, allowing the system to serve its own gradient estimation needs without external assistance or added complexity
Data Source
AI summary
The present disclosure provides an apparatus and method for controlling automatic parking. A method may include obtaining parking line recognition information from a captured image of a target parking slot, determining angular relationships between a plurality of parking lines of the target parking slot based on the parking line recognition information, estimating a gradient of the target parking slot based on the angular relationships, and controlling longitudinal driving of a vehicle based on the estimated gradient.