SBIR Phase I: A Passive Alternative to LiDAR for Automotive 3D Ranging

该小型企业创新研究（SBIR）第一阶段项目的更广泛影响/商业潜力将推动自动驾驶汽车检测系统的开发。所提出的技术利用了智能手机中成像器和处理器的价格，性能和质量的趋势，这些设备的激增推动了这些设备的发展。该技术还提供了关键功能，如集成的颜色信息，扩展深度的检测，场景分割和跟踪，以及在恶劣天气或能见度低的情况下更好的性能。这导致了三维（3D）视觉和对象建模的改进，这将在加速自主和半自主车辆导航和辅助的开发和部署方面产生重大的商业影响。这将导致更高的通勤效率，减少交通死亡人数，减少交通拥堵，减少污染。这个小企业创新研究（SBIR）第一阶段项目将建立基于被动传感图像的多相机EPI Epipolar-Plane Imaging（EPI）分析的技术能力和优势，用于自动驾驶汽车（AV）测距。研究的目的是推进EPI分析的发展，并将其与光探测和测距（LiDAR）系统进行比较。 该研究将扩展现有的基于EPI的模块，以整合新的硬件和软件来实现这些结果，包括在200米及以上的距离上与LiDAR相当的准确度和精度，优于LiDAR的上级特征识别，在所呈现的范围信息中具有更高级别的语义，该奖项反映了NSF的法定使命，并通过评估被认为值得支持使用基金会的知识价值和更广泛的影响审查标准。

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase I project will advance the development of detection systems for autonomous vehicles. The proposed technology takes advantage of trends in price, performance, and quality of imagers and processors driven by the proliferation of these devices in smartphones. The technologyalso provides key capabilities such as integrated color information, detection over extended depths, scene segmentation and tracking, and better performance in inclement weather or under poor visibility. This leads to improvements in three-dimensional (3D) vision and object modeling that will have significant commercial impact in accelerating the development and deployment of autonomous and semi-autonomous vehicle navigation and assistance. This will lead to higher commuting efficiency, reduced traffic fatalities, reduced traffic congestion, and reduced pollution.This Small Business Innovation Research (SBIR) Phase I project will establish the technical capabilities and advantages of passive sensing image-based multi-camera EPI Epipolar-Plane Imaging (EPI) analysis for autonomous vehicle (AV) ranging. The research objective is to advance the development of EPI analysis and compare it to Light Detection and Ranging (LiDAR) systems. The research will extend an existing EPI-based module to incorporate new hardware and software to achieve these results, including accuracy and precision comparable to LiDAR at distances of 200 m and beyond, feature discernment superior to LiDAR, higher levels of semantics in presented range information, and operation in inclement weather and discrete obscuration.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.