STTR Phase I: Versatile Robot Hands for Warehouse Automation

从仓储和物流市场的规模来看，这个项目的更广泛影响/商业潜力是显而易见的。美国劳工统计局估计，每年手工搬运库存的成本为50亿美元。工业界渴望采用新的机器人技术，它已经在这个问题的许多方面产生了重大影响，加快了发货速度，降低了成本。然而，当对象范围很大时，处理单个项目尚未被证明是可行的。本课题开发的抓取机器人将满足这一需求，降低制造商、分销商和客户的物流成本。此外，能够自动抓取各种物体的机器人抓取系统的创建，将为其他行业的工人带来机器人技术的好处。这对于使小型制造车间能够使用机器人至关重要，并将显著提高用于军事、救援和消费应用（如老年人家庭辅助）的远程呈现系统的性能。这项小型企业技术转移研究（STTR）第一阶段项目是基于大学十年来对使用被动机构抓取机器人的设计和构造的研究。通过精心调整手指的结构顺应性，可以设计机器人的手来补偿被抓取物体的大小、形状和位置的变化，以获得可靠的抓取。在这个项目的过程中，一种商业产品将被开发出来，能够以极低的错误率挑选和放置各种各样的物品，这在传感和规划方面一直被认为是一个难题。新型兼容的抓取器，结合低成本触觉传感和简化抓取规划方面的突破，将实现满足现实世界客户需求的性能水平。新的拾取机器人将在模拟的仓库环境中对从垃圾箱中拾取的一组真实物品进行验证。

The broader impact/commercial potential of this project is clear from the size of the market in warehousing and logistics. The Bureau of Labor Statistics estimates that the annual cost of moving inventory by hand in warehouses is 5 billion dollars. Industry is eager to adopt new robotic technology, and it has already made a significant impact in many aspects of this problem, speeding shipments and lowering costs. However, handling individual items has not yet proven viable when the range of objects is large. The grasping robot developed in this project will meet this need, and reduce the cost of logistics for manufacturers, distributors, and customers. Additionally, the creation of robot grasping systems that can automatically grasp a wide range of objects will open up the benefits of robotic technology to workers in other industries. This is critical to making robots accessible to small manufacturing shops, and will significantly improve the performance of telepresence systems used for military, rescue, and consumer applications such as home assistance for the elderly.This Small Business Technology Transfer Research (STTR) Phase I project is based on a decade of university research on the design and construction of robots for grasping using passive mechanisms. Through the carefully tuned structural compliance of the fingers, robot hands can be designed to compensate for variation in the size, shape and location of grasped objects to obtain reliable grasps. During the course of this project, a commercial product will be developed capable of picking and placing a wide range of items at extremely low error rates, something that has hitherto been considered a hard problem in sensing and planning. The new compliant grippers, combined with breakthroughs in low-cost tactile sensing and simplified grasp planning, will enable a level of performance that meets the needs of real-world customers. The new picking robot will be validated against a set of realistic items picked from bins in a simulated warehouse environment.