Integrated LiDAR system for autonomous vehicles

适用于自动驾驶汽车的集成激光雷达系统

• 批准号: 534012-2018
• 负责人: Ye, Winnie
• 金额: $ 2.91万
• 依托单位: Carleton University
• 依托单位国家: 加拿大
• 项目类别: Collaborative Research and Development Grants
• 财政年份: 2021
• 资助国家: 加拿大
• 起止时间: 2021-01-01 至 2022-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=719969
• 关键词: Integrated; LiDAR; system; autonomous; vehicles

LiDAR (Light Detection and Ranging), is a surveying method that measures distance to a target by illuminatingthe target with pulsed laser light and detects the reflected pulses with a sensor. If costs can be lowered andfunctionality improved, the market for LiDAR in automobiles and unmanned vehicle is expected to surpassUSD$2.5 Billion by 2030. An approach to this challenge is to create "fast agile" LiDAR platforms based onphotonic integrated circuits, integrated with advanced Artificial Intelligence (AI) systems. Such systems couldsupport modules that would "attend to" objects of importance in the field of view allowing for high-resolutionidentification of critical items. LiDAR will not be simply a passive information gathering element but anactively controlled part of an integrated sensing aware AI platform.This grant has two overarching goals. Firstly, it proposes to generate a viable PIC prototype for theimplementation of a "fast agile" LiDAR with the necessary functionality to enable an "attention module" withinan AI infrastructure. Secondly, this project proposes to develop a sophisticated simulation platform for PICbased LiDAR. The proposed research will form a foundation for the complete system level design of theLiDAR PIC.The first goal will improve the functionality of the LiDAR by employing depletion-based phase shifter arraysand subwavelength grating couplers to improve beam steering range, resolution and scanning speed. Thesecond goal will be complimentary to the first by providing analysis and optimization capabilities. This workwill address the multi-physics and multi-scale issues present in LIDAR design and optimization, dealing withsystems by incorporating electrical, optical, semi-conductor and thermal devices and effects. The impact of thephysical layer on system design and constraints will also be addressed. The simulation work will build on theexisting simulation frameworks of OptiSPICE and Atar. Experimental verification of the modeling usingprototype devices are anticipated. Two prototype generations are expected by the end of the CRD program.

激光雷达（Light Detection and Ranging）是一种测量方法，它通过用脉冲激光照射目标并用传感器检测反射的脉冲来测量到目标的距离。如果能够降低成本并改善功能，到2030年，汽车和无人驾驶汽车中的LiDAR市场预计将超过25亿美元。应对这一挑战的方法是创建基于光子集成电路的“快速敏捷”LiDAR平台，并与先进的人工智能（AI）系统集成。这样的系统可以支持模块，将“照顾”的重要对象在视野中允许高分辨率识别的关键项目。LiDAR将不仅仅是一个被动的信息收集元件，而是集成传感感知AI平台的主动控制部分。首先，它建议生成一个可行的PIC原型，用于实现具有必要功能的“快速敏捷”LiDAR，以在AI基础设施中启用“注意力模块”。其次，本计画提出发展一个先进的PIC激光雷达模拟平台。本文的研究将为激光雷达PIC的系统级设计奠定基础。第一个目标是通过采用基于耗尽的移相器阵列和亚波长光栅耦合器来改善激光雷达的功能，以提高光束控制范围、分辨率和扫描速度。第二个目标将通过提供分析和优化能力来补充第一个目标。这项工作将解决激光雷达设计和优化中存在的多物理场和多尺度问题，通过整合电气，光学，半导体和热器件和效应来处理系统。物理层对系统设计和约束的影响也将得到解决。仿真工作将建立在现有的OptiSPICE和Atar仿真框架上。使用原型设备的建模的实验验证是预期的。两代原型预计在CRD计划结束时完成。