SBIR Phase I: Path Planning for Multi-target Search and Localization in Co-Robotic Architectures

这个小型企业创新研究(SBIR)第一阶段项目的更广泛/商业影响是一种经济且易于使用的高性能、基于无人机的野生动物遥测技术。其改进的性能和比现有竞争技术低得多的成本将赋予用户更大的能力，并降低实地数据收集的进入门槛。这项技术将使研究人员能够更好地了解地理、气候、物种间竞争、入侵物种和土地利用政策对动物及其栖息地的复杂影响。该项目将实现一个联合机器人系统，该系统使用无人机作为辅助野生动物追踪器。通过融合无人机的数据和人类跟踪器的知识和洞察力，该团队将极大地提高这些系统跟踪更多动物的能力，增加跟踪活动的频率，并提高易用性。这项研究将开发和提炼一类新的用于野生动物定位的机器学习算法。随着无人机平台执行其任务，人类跟踪器可以监控其进度和计划，并随时通过命令干预无人机重新定向。该研究计划解决了这一问题领域的关键挑战，包括无人机操作的安全和噪音排放可能产生的影响。在确定关键风险和缓解策略的基础上，研究计划在一系列任务中执行开发-模拟-部署-测试设计流程的多次快速迭代，以应对日益复杂的本地化场景。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

The broader/commercial impact of this Small Business Innovation Research (SBIR) Phase I project is a high-performance, drone-based, wildlife telemetry technology that is economical and easy to use. Its improved performance and significantly lower cost than existing competitive technologies will empower users with greater capability and lower the barrier to entry for field data collection. This technology will enable researchers to better understand the complex effects of geography, climate, interspecies competition, invasive species, and land use policy on animals and their habitats. This project will enable a co-robotic system that employs an unmanned aerial vehicle (UAV) as an assistant wildlife tracker. By fusing data from the UAV and the knowledge and insights of the human tracker, the team will greatly increase the capability of these systems to track a larger number of animals, to increase the frequency of tracking campaigns, and to improve the ease of use. This research will develop and refine a new class of machine learning algorithms for wildlife localization. As the UAV platform executes its mission, the human tracker can monitor its progress and plans, and intervene at any time with commands to re-direct the UAV. The research plan tackles the key challenges in this problem domain, including safety of UAV operation and the possible effects of noise emissions. Based on the identification of key risks and mitigation strategies, the research plan executes multiple rapid iterations of a develop-simulate-deploy-test design process within a set of tasks that tackle increasingly complex localization scenarios.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.