Autonomous Vehicle Path Planning Using Map Data and Sensor Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current vehicle path planning systems for semi-autonomous or autonomous vehicles face limitations in accurately determining lane centering and lane changing paths due to sensor and actuator latency, and the limited effective viewing distance of vision cameras, which can lead to improper path generation and harsh maneuvers, especially at higher speeds.
Innovation Solution
A system and method that utilize roadway points from a map database to determine a reference vehicle path, combined with sensor data for obstacle detection and lane marking recognition, to generate a reduced curvature path and provide multiple candidate paths and speeds for safe navigation, including the use of fifth-order polynomial equations for path modeling and segmentation of the roadway into multiple segments based on detection range and curvature.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If vision cameras are used to detect roadway markings for path generation, then lane centering and lane changing control can be achieved, but the limited effective viewing distance of about 80 meters restricts the ability to detect roadway markings and determine roadway curvature, especially at high speeds
Solution Approach 1:
The patent combines multiple data sources including vision camera data, map database information, and vehicle sensor data to determine roadway curvature and generate reference paths. This merging compensates for the limited camera viewing distance by incorporating map data that provides curvature information beyond the camera's 80-meter range.
Solution Approach 2:
The patent introduces map database information as an intermediary to bridge the gap between the camera's limited viewing distance and the required detection range for accurate curvature determination. The map data serves as a mediator that provides curvature information where direct camera detection is insufficient.
2Productivity
If sensor and actuator latency is present in the system, then real-time vehicle state measurement is achieved, but the measured vehicle states differ from actual vehicle states, causing improper path generation and harsh maneuvers
Solution Approach 1:
The patent generates a reference path in advance using map data and curvature information, then uses this pre-computed reference path to guide real-time vehicle control. This preliminary path generation compensates for latency by having the desired trajectory ready before execution begins.
Solution Approach 2:
The patent implements feedback mechanisms where the system continuously compares actual vehicle states with the reference path and adjusts control commands accordingly. This feedback loop helps correct deviations caused by sensor and actuator latency, maintaining path following accuracy despite real-time measurement errors.
3Speed
If the camera range is insufficient to complete a lane change maneuver at highway speeds, then the system can detect objects within 80 meters, but the vehicle speed exceeds the system's ability to accurately provide the necessary predicted path
Solution Approach 1:
The patent pre-generates reference paths using map data and curvature information before the vehicle reaches the decision point. This allows the system to have accurate path information ready in advance, compensating for the limited time available at high speeds to process sensor data and compute trajectories.
Solution Approach 2:
The patent merges map database information with real-time sensor data to create a comprehensive view of the roadway geometry and curvature. This combination provides accurate predicted path information even when the camera's real-time detection range is insufficient for high-speed maneuver planning.
Data Source
AI summary
A method for automated lane centering and/or lane changing purposes for a vehicle traveling on a roadway that employs roadway points from a map database to determine a reference vehicle path and sensors on the vehicle for detecting static and moving objects to adjust the reference path. The method includes reducing the curvature of the reference path to generate a reduced curvature reference path that reduces the turning requirements of the vehicle and setting the speed of the vehicle from posted roadway speeds from the map database. The method also includes providing multiple candidate vehicle paths and vehicle speeds to avoid the static and moving objects in front of the vehicle.