Insects’ acrobatic and resilient flight traits help them navigate wind gusts, obstacles, and general uncertainty. Such traits are hard to build into flying robots, but Massachusetts Institute of Technology (MIT) Assistant Professor Kevin Yufeng Chen built a system that approaches insect agility.

Chen, a member of the Department of Electrical Engineering and Computer Science and the Research Laboratory of Electronics, developed insect-sized drones with dexterity and resilience.

Typically, large drones aren’t nimble enough to navigate confined spaces nor robust enough to withstand collisions. Tiny drones require a different construction from larger ones, since motors lose efficiency as they’re miniaturized. For insect-like robots, the principal alternative has been a small, rigid actuator built from piezoelectric ceramic materials. While piezoelectric ceramics allowed the first generation of tiny robots to take flight, they’re extremely fragile – a problem when building a robot to mimic a foraging bumblebee that endures a collision about once every second.

Chen designed a more resilient tiny drone using a new class of soft actuators instead of hard, enabling the aerial robots to withstand the physical struggles of real-world flight. The soft actuators are made of thin rubber cylinders coated in carbon nanotubes. When voltage is applied, the carbon nanotubes produce an electrostatic force that squeezes and elongates the rubber cylinder. Repeated elongation and contraction cause the drone’s wings to beat rapidly.

Chen’s actuators flap nearly 500x per second, giving the drone insect-like resilience. “You can hit it when it’s flying, and it can recover,” Chen says. “It can also perform somersaults in the air.”

The current prototype weighs 0.6g, approximately the mass of a large bumble bee. Chen is working on a new model shaped like a dragonfly. A key step will be untethering the drone from a wired power source, which is currently required by the actuators’ high operating voltage.

Chen says his mini-aerialists could navigate complex machinery to ensure safety and functionality.

“Think about the inspection of a turbine engine. You’d want a drone to move around [an enclosed space] with a small camera to check for cracks on the turbine plates.”

Other potential applications include artificial pollination of crops or completing search-and-rescue missions following a disaster. “All those things can be very challenging for existing large-scale robots.”

Massachusetts Institute of Technology (MIT)

Chen’s work, “Collision Resilient Insect-Scale Soft-Actuated Aerial Robots with High Agility,” appears in the journal IEEE Transactions on Robotics. Co-authors include MIT Ph.D. student Zhijian Ren, Harvard University Ph.D. student Siyi Xu, and City University of Hong Kong roboticist Pakpong Chirarattananon.

Collaborative robots (cobots) assist human workers and can handle extraordinary circumstances, even a novel global pandemic. As the technology changes and expands capabilities, cobots have become increasingly important due to growing demand for enhanced productivity, speed, and safety. According to the latest report by Research and Markets, the cobot market is projected to grow at a compound annual growth rate (CAGR) of 30.37% from 2020 to 2025. Kristian Hulgard, general manager – The Americas at OnRobot, examines factors influencing this growth and poses expectations for what’s ahead.

Aerospace Manufacturing and Design (AM&D): What value is created with collaborative operations, and what are the advantages of having these systems in place?

Kristian Hulgard (KH): For us, it’s about creating value in the entire supply chain. It’s not only about creating value for the end user, but also for the integrators and distributors. Automation is easy to use for quick and effective deployment with shorter integration time and lower costs. In addition, because it’s flexible, the manufacturer can manage, operate, and maintain a program on their own.

Collaborative operation is also addressing current challenges in the market. It solves labor shortages, improves product quality, increases productivity, provides fast return on investment (ROI), and helps with social distancing to keep workers safe. It was very popular in the 1990s to outsource manufacturing to Asia; now collaborative operations help with the changing focus toward reshoring.

AM&D: The market is heavily focused on software-driven solutions. Where are you seeing this shift and why?

KH: Traditionally, components on a robot and integrated in an application cell were pneumatically driven, so it was either on or off. Those were the only two options. Now demand from the market is leaning more toward electric tools and accessories controlled by software. Instead of end of arm tooling (EOAT) having only one function, they are more flexible and have multiple capabilities. For example, a gripper doesn’t only get one part, the gripper can grip different shapes, sizes, and materials. And with more Industry 4.0 capabilities in place, manufacturers can monitor their manufacturing in the cloud, determine maintenance ahead of time, and monitor machines to see how the robot is running and its effectiveness. If you have software driven tools, sensors, and cameras, then you can export that and optimize your production, meaning you’re more productive and profitable.

AM&D: What is the market outlook for cobots, and what trends are you seeing?

KH: Like the rest of the world, we were of course challenged by COVID-19. As many manufacturers made shifts in production and found new use cases for cobots, newcomers to automation demonstrated more growth compared to the more traditional cobot and light industrial robot industries. Looking into 2021, I believe we’ll see more traditional automation users come forward again.

AM&D: How will application design change and advance?

KH: There are many more companies taking the software and the plug-and-play approach, and I think that’s the future. You integrate a part onto the robot, connect it to the controller, and you’re ready to use it instantly. I can also see the difference as more manufacturers are partnering to develop these application capabilities. I would say in 5 to 10 years, for lightweight, industrial and collaborative robot application design, it’ll be something you can do online, through drag and drop, put all your components in something that will generate what you need, and then send it to your local distributor who will ship you the parts.

I also think that data and machine learning will involve simpler, almost fully automatic application programming. Light industrial robots will be standard components in the collaborative operation. The focus will shift more to how easily and quickly the solution can solve a problem.

AM&D: Why is developing one language for all robots so important for the future?

KH: We want one language and one software for all robots so we can open automation to everyone. It’s very important to look at components and parts for the application, where you have just one control or one language. Compatibility is usually a huge barrier, especially for small- and medium-sized enterprises (SMEs) when they don’t have a complete overview of all the components and each part has its own language. It’s critical to have one single common platform so that we can give users the safety and confidence they need to use any tool.

AM&D: What do you anticipate moving forward in this market?

KH: I think there’ll be less talk about specific gripper or robot functions, and we’ll be focusing more on solutions, but how are we going to get there as fast as possible? With easily deployable robots.

The International Federation of Robotics states that we’re going to sell 10x as many robots in five years. And my question to the industry is, how are you going to contribute to 10x as many robots deployed? The answer is not going to be hiring 10x as many engineers or salespeople. It’s determining how to deploy a robot or automation solution faster and more effectively. In the time that I spend deploying one robot today, in five years, I can deploy 10 robots. That’s key to the market’s success, and there are several companies contributing to that. Eliminating deployment time and decreasing hours on design, installation, and programming is essential.

OnRobot

All photos courtesy of Avantor NuSil

The U.S. Air Force discovered a problem in the 1960s: silicone-based sealants, coatings, and other materials used around fuel-exposed surfaces required more frequent repairs or failed prematurely.

Naturally soluble, standard silicones (dimethyl silicones) easily absorb fuels, crude oil, de-icing fluids, and other hydrocarbon-based solvents, causing the materials to swell, which weakens the silicone network, diminishes mechanical properties, and distorts dimensional shape. Swelling can cause expansion and cracking that breaches a seal or gasket, and that degradation can generate unplanned downtime, increased maintenance, and even catastrophic failure.

The Air Force required a material that could withstand cold high-altitude environments, engine exhaust heat, and UV exposure without sacrificing performance. While dimethyl silicone offers thermal stability in many applications, prolonged exposure to sunlight can cause cracking.

Fluorosilicones were the solution. Less soluble to fuels and organic solvents, they resist swelling while maintaining their physical properties and dimensional shapes. The materials better resist mechanical failures caused by hydrocarbon exposure. Additionally, fluorosilicones remain flexible at extremely low temperatures and resist breakdown at very high temperatures.

With decades of flight service, fluorosilicones provide reliable protection from breakdown, even against prolonged exposure. Widely used across various military and civilian aircraft components, their versatility makes them ideal for:

• Adhesives and sealants for wire staking, bonding components, sealing joints with different coefficients of thermal expansion; attaching composite materials to aircraft exteriors
• Coatings, sheeting for aircraft outer mold line, fuel cells; thin coatings on windshields, other surfaces
• Electrically conductive gap fillers for small cavities between aircraft surfaces
• Gels, foams for encapsulating sensitive electronics for vibration, shock; potting
• Molded gaskets, bushings used in or near engine compartments, fuel tanks

Available in a variety of forms and cure chemistries, fluorosilicones are suitable as molding compounds, paints, coatings, and one- or two-part adhesives. This versatility improves aircraft manufacturer processing and application efficiency while providing options based on end-use chemical and physical requirements.

An experienced silicone technology partner can customize formulations to include additional functional fillers, adjustable cure times, and other processing requirements. Formulations can also be customized with color matching to AMS-STD-595 or Pantone Matching System.

Evaluating fluorosilicones

Because fluorosilicones must meet stringent aircraft operating environment requirements, it’s critical to properly validate functions when developing specific solutions. Fluorosilicones should be evaluated against MIL DTL 25988C, which outlines elastomer testing requirements. Softer fluorosilicones may not have the same mechanical strength as harder fluorosilicones, but they will still have excellent temperature and fuel resistance.

Testing an elastomer’s performance can ensure stability while resisting degradation from harsh temperatures and fuel exposure.

Thermal stability: The elastomer’s ability to resist mechanical breakdown, thermal stability ensures the fluorosilicone can function properly during operation as well as reduce downtime and repair. Thermogravimetric analysis (TGA) is widely accepted to evaluate the thermal stability of silicones and fluorosilicone polymeric materials. It evaluates changes in mass when exposed to incremental temperature changes.

Weight loss: Conducted in a convection oven, this test evaluates how the fluorosilicone reacts when exposed to high-heat conditions for an extended time.

Swelling: The swelling test measures the percentage of mass change when a material absorbs hydrocarbons.

Other considerations

The key to efficiently identify the right fluorosilicone is to collaborate with an experienced silicone technology partner. It’s critical that the formulator possesses deep expertise in each unique application and key performance requirements. It’s also important that a provider complies with all relevant regulations and has an established aerospace heritage.

Fluorosilicones are ideal for mission-critical aircraft systems in which failure is not an option. Understanding these elastomers, how the materials perform, and what to consider when developing a solution can help pinpoint the right solution in aircraft applications.

Avantor NuSil

About the authors: Michelle Velderrain is senior product and application specialist, and Robert Umland is director of advanced technologies, NuSil – part of Avantor. They can be reached at 805.684.8780 or silicone@nusil.com.

Longtime relationships with its employees, customers, and suppliers, along with access to real-time manufacturing data, are helping a Connecticut aerospace job shop survive the disruptions of the COVID-19 pandemic and prepare for renewed growth in its wake.

When the pandemic appeared, PDQ immediately took measures to protect its employees, including regular body temperature checks. Slowing or stopping production weren’t options as a military aerospace parts customer named the shop an essential business.

“We are to remain open and fulfill deliveries as needed, and that’s where we are right now,” Manufacturing Manager Scott Norman says.

Machine data challenges

PDQ was undergoing a comprehensive enterprise resource planning (ERP) software implementation when the COVID-19 crisis began, making part demand and data management more volatile. For a company that already needed improvements to machining data recording, it was a challenging time.

“In the old system, operators wrote their setup and run times on paper time sheets,” Norman explains. “The production data was entered at the end of the day and wasn’t immediately verifiable. We were a slave to when and how the information was put in.”

Delayed and sometimes inconsistent manual recording of machine output resulted in shop personnel not always being sure of finished part volume and expected lead time.

As LillyWorks (Hampton, New Hampshire) installed the ERP, PDQ officials decided to integrate the software with a machine data communications (MDC) platform from Excellerant LLC (Somers, Connecticut). Excellerant’s software and hardware act as a universal decoder, allowing a shop to connect various brands of CNC machine tools and controllers, monitor and manage each machine’s data in real time, and then transmit the data to the company’s ERP system. Excellerant communicates via MTConnect, Fanuc Focas, OPC-UA, Haas MNET Q-Codes, and other machine control connecting protocols and can be used with legacy CNC machines.

Excellerant’s system counts parts as they’re created. At the end of each shift, the operator qualifies the number of good parts produced and the MDC platform automatically enters the total into the shop’s ERP system.

Norman says using the Excellerant platform has made it easier for his staff to do its job. “I don’t have to leave my office to tell operators what needs to be done. They know what they should be doing, how many parts they should be making. Previously, that wasn’t communicated clearly to them, now it is.”

PDQ President and CEO Ron Gronback Jr. adds that with the automated system, “You turn the machine on and Excellerant knows every button you push. Turn it on and not run it, it knows that, too.”

“Every machine can be monitored wirelessly,” Norman says. “We use Excellerant for our DNC, for labor reporting, and for tracking part counts, uptime, and downtime. I have a 55" monitor on the wall outside my office that everyone can see. All the machines are highlighted in either yellow, green, or gray, with yellow indicating that a machine is idle, green that it’s running, and gray that there’s no connection or it isn’t turned on.”

Real-time machine data monitoring helps PDQ maximize spindle use.

“The more jobs you run through the system, the more representative the data becomes and that makes it much better,” Norman says. “The accurate, real-time data confirmed things that Ron and I knew all along. It wasn’t about finding out who is working and who isn’t. Excellerant gives us real data for better planning. Now, I can quote with accurate setup times, etc.”

Because Excellerant is integrated with the LillyWorks ERP, “when an operator puts in the proper information, it will trigger the materials replacement plan if we’re short.”

Excellerant also works with LillyWorks’ Protected Flow Manufacturing (PFM) feature, a predictive process tool that analyzes how parts are being made, machine efficiency, and the supply chain for hardware and raw materials. It then determines the order parts should be processed to maximize quality, efficiency, and on-time delivery. Algorithms consider the upcoming operations and indicate which job should be worked on next – it’s not always those due soonest.

Operator-friendly machine

Excellerant gives operators a way to report and document machine stoppages for part measurements, broken tools, or other interruptions. An operator can assign reason codes to times when the spindle’s not turning, and the office can analyze that input to determine downtime causes.

Rick Lees, Excellerant’s director of technical operations, says it’s important to involve the machine operators so they know how automated data acquisition can make their jobs easier. “If you tell the operators ‘here’s what we’re doing,’ it can get people’s guard up. But if you present the system as a way the operators can tell the office why a machine is idle, what element of the process resulted in a slowdown or shutdown, it provides instant and solid documentation for any problems.

“We’ve made Excellerant a more useful tool for PDQ by integrating it with their ERP system. When the PDQ front office receives an order, it enters the job into LillyWorks and assigns it to a machine or machines on the shop floor. Excellerant follows the job and displays the work order, the operation, and the number of parts needed so the operator knows what to do. As the job progresses to completion, the time data comes directly from the machine control. The operator doesn’t have to remember how much time was spent on that part.”

Operators can also see how many parts they should have at any time, how many parts they’ve made, and what’s next.

“They can see what’s next on the list and point out any concerns. If they say, ‘I have all the tools in this machine already, it will be a really quick changeover,’ it winds up being a real timesaver. This doesn’t work without the operators being in the loop,” Lee says.

Appreciated

Norman underlines the contributions of Excellerant staff to the success of the installation, including Lees, senior software architect Casey Marquis, and lead service technician Michael Paul Loiselle. “They listened. I can tell they are listening to other customers too, because since we started with Excellerant, they’ve updated it a few times.”

Lees says, “PDQ is in a very positive spot with Excellerant. Scott is an asset to making the product better because he’ll write us and ask, ‘Hey, can it do this, or can it do that?’ Customer use is the best tool to make the product better. The things we’ve added at their request have enhanced the system. We’re really partnering on this. We’re not just handing them devices and saying, ‘Here you go, good luck.’ You know, it’s all about the relationships.”

Excellerant LLC

LillyWorks

PDQ Inc.