SBIR-STTR Award

The broader impact/commercial potential of this SBIR Phase I project is to reduce the time required to construct concrete bridges, improve job site safety relating to rebar installation, reduce costs related to bridge construction, and improve the overall health of ironworkers. Ironworkers face some of the highest rates soft tissue damage of the industrial occupations due the constant bending over required to tie rebar. Additionally, by reducing the required construction time for bridges, communities will be able to recover from natural disasters at a faster pace. These benefits will be accomplished by automating the process of the tying rebar. This is significant due to the highly repetitive nature of the tying process, the labor shortages in the construction industry, and the fact that rebar tying often sits on the critical path of a concrete pour. To automate rebar tying, a drone platform with an integrated tie tool, specialized flight controls, and navigation system will be developed. This will take small unmanned aircraft systems (sUAS) from observation roles to a manipulation platform. Rebar tying on bridges represents a $275 million market over the next 10 years and $1.7 billion market annually across the USA for the larger general construction market. This Small Business Innovation Research (SBIR) Phase I project aims to develop a sUAS capable of tying the rebar for concrete construction on an outdoor fixture. This will represent the first commercially viable aerial manipulation system if the project is successful. A combination of computer vision, machine learning, and sensor fusion techniques will be employed to develop an autonomous system capable of allowing a drone to identify, land and tie with a high level of accuracy to enable rebar tying. Modification to the autopilot will be conducted to allow the drone to trigger the integrated rebar tool. Computer vision algorithms will be optimized and ported to run onboard the sUAS. A control system for visual servo-ing will be developed to utilize onboard computer vision algorithms and other sensors to track and land on rebar intersections accurately. High-level challenges consist of uncovering the robustness of existing rebar tool towards imprecise landing, improving landing precision, improving accuracy of rebar intersection detection, and visual servo-ing of the drone. 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.

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase II project is the development of a robotic solution to automate a dangerous and repetitive rebar tying task in construction. Rebar tying is performed by “rodbusters”, 88% of whom reported that they suffered some form of musculoskeleton disorder during their career. Removing the repetitive bending over and tying actions can have significant impacts on the quality life of the workers who commonly suffer from carpal tunnel complications and back injuries. The market for rebar tying services solely for bridge decks in the US is valued at over $370 million across the next ten years. The general construction market is estimated to spend at least $1.3 billion annually for rebar tying. A quality transportation infrastructure has direct impacts on national defense readiness, economic competitiveness, disaster response, exchange of goods, and standards of living. However, at present, the US has a growing backlog of bridges deemed deficient. A successful Phase II project may play a key role in addressing the infrastructure maintenance backlog and allowing a key foundation of the national economic engine to function properly.This Small Business Innovation Research (SBIR) Phase II project seeks to demonstrate multiple robots to perform automated rebar tying. Many construction projects have thousands of rebar intersections that need to be tied; A system to detect these intersections is proposed. Various algorithms for detecting rebar will be improved and integrated to build a semi-automated rebar map building software. The research project will build an aircraft configuration and develop various computer vision and controls algorithms to demonstrate that a drone can fly over placed rebar, identify the intersections to tie, land on top of them and tie the rebar together. A small swarm of the customized drone prototypes will be built for demonstration. Speed in executing the job is critical for most construction projects implying that each intersection has to be tied in a matter of seconds for viable adoption of the proposed technology. The success of this project may result in the first commercially-viable swarm of mobile manipulation robots for unstructured environments.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.