SBIR Phase I: Sonar Arrays for Maximizing Aquaculture Yields

这项小企业创新研究（SBIR）项目的更广泛影响/商业潜力将是一种利用声纳和分析软件改善虾养殖的方法。价值450亿美元的捕虾业对池塘里有多少虾知之甚少。农民采用“最佳猜测”式的方法，利用不完整且往往不准确的信息做出日常决策，导致饲料到生物质的转化率较低，收获时间不理想以及各种供应链可追溯性和管理问题。本项目的目标是开发一种基于声纳的无创评估系统，以提高原位估计虾的丰度和生长的准确性和便利性。拟议的系统将为农民提供可操作的数据，导致以下价值主张：降低饲料转化率（饲料滴定），提高盈利能力（确定收获时间），会计控制（多个站点的农民控制生产），最大化生长条件（放养密度），减少风险因素（生物安全），并采取紧急行动（作物分类）。该SBIR一期项目建议开发一个传感器和分析平台，用于实时估计虾的生物量和行为，为虾类养殖的决策和最佳实践提供信息。通过优化一系列网络水下声纳传感器，可以在高度曝气的养殖环境中非侵入性地捕获特定物种的虾丰度。此外，算法将为最终用户提供更好的丰度、大小、增长和行为异常的估计，并为进行/不进行决策提供信息，同时降低与不确定性相关的沉没成本。为农民、买家和科学界提供“大数据”。水产养殖为研究提供的受控环境是一种未开发的资源，因为数据没有在池塘、区域或国家之间标准化。拟议技术的发展将是通过同行评议的出版物向公众传播有关传感器优化、动物生长、生产、行为和疾病的重要研究的关键。虽然本研究和本产品的范围主要集中在虾类领域，但该模型可以扩展到其他养殖物种和应用（即鲑鱼和罗非鱼）。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) project will be a method to improve shrimp farming using sonar and analytical software. The $45 billion shrimp industry operates with little understanding of how many shrimp are in a pond. Farmers take a "best guess" type of approach using incomplete and often inaccurate information to make daily decisions that results in poor feed to biomass conversion rates, suboptimal harvest timing and a variety of supply chain traceability and management issues. The goal of this project is to develop a noninvasive sonar-based assessment system to improve the accuracy and convenience of estimating shrimp abundance and growth in-situ. The proposed system will provide the farmer with actionable data that lead to the following value propositions: reducing the feed conversion ratio (feed titration), increasing profitability (determine harvest time), accounting controls (farmers with multiple sites controlling production), maximizing growth conditions (stocking density), reducing risk factors (biosecurity), and taking emergency action (crop triage). This SBIR Phase I project proposes to develop a sensor and analytics platform that would allow for estimation of shrimp biomass and behavior in real-time, informing decision making and best practices in shrimp aquaculture. By optimizing a series of networked underwater sonar sensors, it will be possible to non-invasively capture the species-specific abundance of shrimp in the challenging of environment of a highly aerated farm. In addition, algorithms will produce improved estimates of abundance, size, growth and behavioral abnormalities to the end user and informing go/no-go decisions as well as reducing sunk costs associated with uncertainty-based inefficiencies. Making "big data" available to the farmer, the buyers, as well as the scientific community. The controlled setting aquaculture provides for research is an untapped resource because the data are not standardized across ponds, regions or countries. The development of the proposed technology would be key in disseminating important research to the public via peer-reviewed publications on sensor optimization, animal growth, production, behavior and disease. Though the scope of this research and this product focuses on the shrimp sector, the model can be extended to other cultured species and applications (i.e., salmon and tilapia).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.