Self-Driving Route Planning From LiDAR Point Clouds Without HD Maps
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing self-driving vehicle route planning systems rely on high-definition map information and GPS-based navigation, which are costly, require significant manpower, and fail in environments without traffic lane markings, such as intersections and parking areas.
Innovation Solution
A route planning system for self-driving vehicles that uses lidar sensors to detect environmental features and convert point cloud data into aerial views, recognizing traffic lane boundaries and objects without relying on high-definition maps or GPS. The system calculates lane centers, predicts the route of front vehicles, and plans the final route of the host vehicle using either the front vehicle as a reference or traffic lane boundaries.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high-definition map information is used to obtain optimized route, then accurate route planning is achieved, but cost and data storage requirements increase significantly
Solution Approach 1:
The patent extracts only the essential route planning functionality from the high-definition map system by using lidar to directly detect and recognize road features (lane markings, intersections, buildings) in real-time, eliminating the need to store and process massive pre-collected map data while maintaining accurate route planning capability
Solution Approach 2:
The patent creates a real-time digital copy of the road environment through lidar point cloud data and aerial view generation, replacing the need for pre-existing high-definition map copies. The system dynamically generates and updates route information based on current environmental detection rather than relying on static map data
2Measurement precision
If high-definition map information is used to obtain optimized route, then accurate route planning is achieved, but manpower and money spent increase
Solution Approach 1:
The patent implements self-service by enabling the vehicle's lidar system to autonomously detect, recognize, and map road features in real-time without requiring external teams to collect and maintain high-definition map data. The vehicle serves its own navigation needs through self-contained environmental perception and route planning
Solution Approach 2:
The patent replaces the mechanical process of manual map collection and maintenance with an automated optical detection system. Lidar technology automatically captures and processes road environment data, substituting human labor and complex data management infrastructure with a streamlined sensor-based system
3Device complexity
If traffic lane marking detection is used to obtain optimized route, then route planning is simplified, but functionality is lost in environments without lane markings
Solution Approach 1:
The patent creates a universal route planning system that functions across diverse environments by using lidar to detect multiple types of road features (lane markings, intersections, buildings, parked vehicles) rather than relying solely on traffic lane markings. The aerial view generation module synthesizes comprehensive environmental understanding that adapts to various driving scenarios including intersections and parking areas without lane markings
4Measurement precision
If GPS-based navigation is used for route planning, then positioning accuracy is improved, but functionality is lost in areas where GPS fails
Solution Approach 1:
The patent prepares for GPS failure by implementing an alternative positioning and route planning system based on lidar environmental detection and aerial view generation. This cushioning mechanism ensures continuous operational capability in areas where GPS signals are unavailable, such as underground parking garages or urban canyons
Solution Approach 2:
The patent introduces an intermediary system between the vehicle and navigation infrastructure by using lidar to directly perceive and map road features, creating an independent positioning and route planning pathway that does not rely on GPS satellites or external navigation systems
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This solution reduces the cost and data storage requirements associated with high-definition maps, enhances safety by enabling route planning in environments without traffic lane markings, and maintains functionality in areas where GPS fails, such as underground parking.
Implementation Method 1
uses lidar sensors to detect environmental features and convert point cloud data into aerial views
Implementation Method 2
uses the intensities of the return waves to recognize the objects in the environment
Data Source
AI summary
A self-driving vehicle route planning system detects and transforms environment information of a host vehicle into an aerial view including the coordinate information of coordinate points to recognize and mark traffic lane boundaries, traffic lanes, and other vehicles therein, calculates central points of the traffic lanes, and then, calculates a speed of a front vehicle according to its positions and works out a predicted route thereof. If the predicted route is the same as a driving route of the host vehicle, the front vehicle is used as a route reference point to calculate a final route of the host vehicle; if not, the traffic lane boundary is used as a route reference line to calculate the final route. The invention can plan the route merely using a point cloud data, greatly reducing the cost HD map information and storage space.