SBIR Phase II: Orchestration of Multiple Robotic Subsystems into a Commercially Viable Robotic Strawberry Harvesting System

该项目的更广泛的影响/商业潜力包括开发浆果采摘后的筛选和处理，这将提供一种安全有效的方法，从自动收割机的加工流程中去除受污染的浆果。该项目将推进机器人采收的机器视觉技术，并将成为一项关键的使能技术，使人们接受安全有效的自动化采收。该项目针对的市场部门是草莓种植者，尽管该技术也将适用于其他类型的水果和蔬菜种植。该项目的自动化收获技术将缓解草莓种植者长期面临的日益严重的劳动力短缺问题，并确保消费者能够负担得起草莓。填补农业劳动力短缺带来的需求是一个价值10亿美元的商机。这个小型企业创新研究（SBIR）二期项目将开发新的视觉处理和检测方法，这对自动化机器人草莓收割机的使用至关重要。草莓种植者是否接受自动收割技术取决于收割机是否能够将坏浆果从植株中移除，而不让不需要的浆果进入收割机的包装流程，从而可能污染大量的浆果。为了实现这一目标，收割机有必要在从自动采摘到最终包装的多个阶段识别和消除患病、腐烂、受损或感染的浆果。识别要淘汰的浆果的分类方法将非常准确，具有非常高的检测率和低的误报率。所开发的方法将适用于受恶劣室外环境影响的农业机械，以及机器人收获设备上的冲击和振动环境。新的处理过程将被开发出来，将允许对整个浆果进行自动检查，而不会损坏果实或产生受感染浆果交叉污染的风险，这将大大提高自动化草莓加工的技术水平。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 project includes development of berry post-picking screening and handling that will provide a safe and efficient way to remove contaminated berries from the processing stream of an automated harvester. The project will advance machine vision technology for robotic harvesting and will be a key enabling technology leading to acceptance of automated harvesting as safe and effective. The market sector addressed by this project is strawberry farmers, though the technology will be applicable to other types of fruit and vegetable farming as well. The automated harvesting technology advanced by this project will alleviate chronic and worsening labor shortages faced by strawberry farmers and will ensure that strawberries remain affordable and available to consumers. Filling the need created by farming labor shortages is a $1 billion business opportunity.This Small Business Innovation Research (SBIR) Phase II project will develop new vision processing and inspection methods vital to enabling use of an automated, robotic strawberry harvester. Acceptance of automated harvesting technology by strawberry farmers hinges on the ability of the harvester to remove bad berries from the plant without allowing the undesirable berries from entering the harvester packaging stream and potentially contaminating large quantities of berries. To achieve this, it will be necessary for the harvester to identify and eliminate diseased, rotten, damaged, or infested berries at multiple stages in the stream from automated picking to final packaging. The classification method that identifies the berries to be eliminated will be extremely accurate, with a very high detection rate and a low false alarm rate. The methods developed will be suitable for installation on a farming machine that is subject to a harsh outdoor environment as well as the shock and vibration environment found on a robotic harvesting device. New handling processes will be developed that will allow automated inspection of the entire berry without damaging the fruit or creating a risk of cross contamination from infected berries, significantly advancing the state of the art for automated strawberry processing.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.