LiDAR Segment Scan Selection for Timely Self-Driving Maneuvers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Self-driving vehicles face safety issues due to delays in processing LiDAR data, which can be as old as a full scanning cycle, leading to inadequate decision-making in rapidly changing environments, especially when combined with the limitations of 2D camera data and the high cost of faster LiDAR systems and processing equipment.
Innovation Solution
A control system that allows for the rapid provision of needed scan data by determining whether to use previous or upcoming LiDAR data based on time parameters and deadlines, ensuring up-to-date information for decision-making, thereby reducing the reliance on outdated data and potentially reducing system costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the LiDAR system uses a default scanning cycle time of 100 ms, then the system cost is reduced, but the data becomes outdated and unsafe for decision-making at high speeds
Solution Approach 1:
The patent divides the continuous scanning cycle into discrete segments, allowing the system to selectively access and process only the relevant segment data needed for current decision-making, rather than waiting for the complete scanning cycle to finish. This segmentation enables timely access to partial but sufficient data.
Solution Approach 2:
The system performs preliminary scanning and data collection during the scanning cycle, preparing data in advance so that when a decision is needed, the most recent available segment data is already ready for immediate use, reducing the effective delay without requiring a faster overall scanning cycle.
2Loss of time
If faster LiDAR systems and processing equipment are used to decrease scanning cycle time, then data freshness is improved, but the system cost increases substantially
Solution Approach 1:
The system uses partial action by accessing and processing only the necessary segment of scan data rather than requiring complete scanning cycle data. This partial data approach achieves timely decision-making without the need for faster (more expensive) LiDAR hardware or processing equipment.
Solution Approach 2:
The patent changes the parameter of data access timing, allowing the system to use data at different stages of the scanning cycle based on need. By changing when data is accessed and processed within the cycle, the system achieves effective speed improvement without hardware changes.
3Device complexity
If camera data is used instead of LiDAR to reduce cost, then system cost is reduced, but depth information is lost requiring more computationally sophisticated approaches
Solution Approach 1:
The system creates a simplified representation or copy of the essential LiDAR segment data that contains only the critical depth and spatial information needed for the specific decision at hand, allowing cameras or simpler sensors to effectively utilize this condensed spatial data without requiring full LiDAR processing capability.
Data Source
AI summary
A method and system for providing scan data for decision making in a self-driving vehicle are provided. The self-driving vehicle includes a scanning sensor that scans a space around the self-driving vehicle in consecutive scanning cycles. The method includes receiving a request for scan information associated with a requested segment of space. The method determines whether to wait for a next scan of the requested segment and use scan data from the next scan, or to extract and use scan data from a previous scan of the requested segment. The method also includes processing the scan data to generate the scan information for the requested segment, providing the scan information for the requested segment in response to the request, and operating the self-driving vehicle in accordance with a decision made about a maneuver to be performed by the self-driving vehicle based, at least in part, on the scan information.