Autonomous Vehicle Path Planning Using Side-Mounted 3D Sensors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous vehicle navigation systems using generalized object and feature detection processes are computationally inefficient and slow, making them unsuitable for real-time path calculation and obstacle avoidance.
Innovation Solution
A computer-implemented method utilizing side-mounted sensors with non-overlapping fields of view to generate 3D point clouds representing the left and right sides of a vehicle, correlating this data to form a 3D representation of the area between sensors, selecting a traversable path, and forming a travelling path based on this data, even in situations where forward-facing sensor data is lacking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If generalized object and feature detection process is used for navigation and obstacle avoidance, then vehicle operation reliability is improved, but computational efficiency deteriorates
Solution Approach 1:
The patent segments the detection process by using multiple specialized sensors (forward-facing sensor, side-facing sensors, rearward-facing sensor) positioned at different locations on the vehicle. Each sensor captures data from a specific direction, and the system processes these segmented data streams separately before integrating them into a comprehensive path calculation, thereby improving both reliability and computational efficiency
Solution Approach 2:
The patent transitions from traditional 2D sensor data to 3D point cloud data by incorporating depth information from multiple sensors. The side-facing sensors provide lateral dimension data, while the forward and rearward sensors provide longitudinal data, creating a three-dimensional representation of the environment that enables more accurate and efficient path calculation
2Reliability
If generalized object and feature detection process is used for navigation, then navigation robustness is improved, but path calculation speed deteriorates
Solution Approach 1:
The system performs preliminary actions by continuously capturing and pre-processing sensor data from multiple directions simultaneously. The side-facing sensors continuously map the lateral environment, and this pre-acquired data is ready for immediate use when path calculation is needed, eliminating the need for time-consuming on-demand detection and improving real-time response
Solution Approach 2:
The patent merges data from multiple sensors positioned at different locations and orientations into a unified 3D point cloud representation. By combining forward-facing, side-facing, and rearward-facing sensor data, the system creates a comprehensive environmental model that enhances navigation robustness while maintaining calculation speed through integrated processing
Data Source
Figure 1A~1B
Figure 2
Figure 3A~3C
AI summary
The present disclosure relates to a computer implemented method for operating a control system (200) to form a travelling path (308) for a vehicle, where the path determination is based on data generated by a pair of sensors producing three-dimensional (3D) point clouds, where the 3D point clouds respectively provides a representation of a left (L) and a right (R) hand side of the vehicle. The present disclosure also relates to the corresponding control system (200) and to a computer program product.