Programmable LiDAR Pre-Processing for Adaptable Point Cloud Pipelines
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing LiDAR systems lack flexibility and adaptability in updating pre-processing algorithms due to the use of dedicated integrated circuits that harden fixed algorithms, leading to complex and costly chip replacements.
Innovation Solution
A modular LiDAR data processing system with a processing module, storage module, programming module, and post-processing module, incorporating on-chip and off-chip programming units, and a write-back mechanism to enable programmable pre-processing, allowing flexible adjustment and extension of the chip's life cycle.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the pre-processing algorithm is hardened to a dedicated integrated circuit, then the processing speed and reliability are improved, but the adaptability and flexibility are worsened
Solution Approach 1:
The pre-processing algorithm is divided into multiple processing units, each handling specific functions. This segmentation allows individual units to be updated or reconfigured independently while maintaining overall system reliability, resolving the contradiction between fixed hardware reliability and algorithmic adaptability.
Solution Approach 2:
The patent implements a dynamic configuration mechanism where processing units can be selectively enabled or disabled based on different algorithm requirements. This dynamic approach allows the hardened circuit to adapt to different pre-processing algorithms without sacrificing the reliability benefits of hardware implementation.
2Adaptability or versatility
If the pre-processing algorithm is updated, then the adaptability is improved, but the device complexity and cost are worsened
Solution Approach 1:
The patent designs a universal processing platform with multiple reconfigurable processing units that can handle different pre-processing algorithms. This multi-functionality eliminates the need for chip replacements when updating algorithms, as the same hardware platform can be reconfigured through software to implement different algorithms, thereby improving adaptability without increasing device complexity.
3Speed
If all data processing is performed on-chip, then the processing speed is improved, but the power consumption is worsened
Solution Approach 1:
The patent implements a selective processing approach where only necessary data processing operations are performed on-chip, while other operations are offloaded. This partial action on-chip maintains high processing speed for critical operations while reducing overall power consumption by avoiding unnecessary computations on the power-hungry chip.
Data Source
AI summary
A LIDAR data processing system includes a processing module, a storage module, a programming module, and a post-processing module, where the storage module includes a first storage unit and a second storage unit, the processing module is used to receive the sampling data, and the post-processing module is used to output the point cloud of the LiDAR. An input port of the first storage unit is connected to an output port of the processing module, and an output port of the first storage unit is connected to an input port of the programming module. An input port of the second storage unit is connected to an output port of the programming module, and an output port of the second storage unit is connected to the post-processing module.