Cambodia has reported more than 64,000 casualties from landmines and explosive remnants of warfare since 1979, and the Southeast Asian nation has the highest ratio of amputees in the world, with more than 25,000 people losing limbs. The number of annual victims in Cambodia has declined to 111 in 2013, but unexploded ordnance still maims and kills hundreds there and in Afghanistan, Columbia, and several other nations.
The U.S., China, India, and Russia have not signed the Convention on the Prohibition of the Use, Stockpiling, Production and Transfer of Anti-Personnel Mines and on their Destruction, informally known as the Ottawa Treaty, but some American roboticists are helping to address the problem.
Although the U.S. military has robots for explosives disposal, “an affordable robot for the developing world isn’t as profitable as a military-grade robot,” said Garrett Clayton, an associate professor of mechanical engineering at Villanova University’s College of Engineering. “There has been some work at universities on this problem. One of the reasons why we’ve been so successful is our partnership with a nonprofit. It has been able to give us specifications for an explosive ordnance disposal [EOD] robot that’s useful but still low-cost.”
Clayton and Villanova University students have been working with the Golden West Humanitarian Foundation on the Low-Cost Explosive Ordnance Disposal Robot. “We bring robotics, the technical side to the equation, and Golden West brings extensive experience in the humanitarian and UXO [unexploded ordnance] remediation community,” he said. “We take groups of mechanical engineers and put them on the problem. This does lengthen the development cycle, but the educational opportunities offered to students are invaluable.”
Happenstance and humanitarian work
The robot’s genesis is the result of chance and very human connections. “A colleague and I were traveling to Cambodia for a service project. Villanova sends students all over the world on humanitarian projects,” recalls Clayton. “We were on a plane from Taipei to Phnom Penh, and one student was having a problem communicating with the flight attendant. The guy sitting next to him helps translate and get his seatbelt fastened.”
“He was a former U.S. Marine EOD tech who invited us to visit his facility,” says Clayton, describing Len Austin, EOD operations chief at Woodland Hills, Calif.-based Golden West. “Golden West had an interest in expanding R&D. We started talking about senior design projects, and their people mentioned the idea of a low-cost way to deal with [UXO].”
Michael Benson, a senior mechanical engineering student at Pennsylvania-based Villanova, learned about the project from fellow students. “They were passionate about making a product that will actually save lives,” he said. “It’s the best thing to choose as a capstone project — a combination of interesting robotics work as well as being able to change lives.”
Over the past three years, a total of 14 students, both undergrads and graduate students, have participated in the project, and several have traveled to Cambodia for field tests. “Being there makes a big difference,” said Clayton.
Trading functions for affordability
“There were three competing goals to trade off when designing a robot like this: functionality, cost, and robustness. Our robot doesn’t do what the PackBot or Talon can do,” Clayton said, referring to iRobot and Foster-Miller’s military machines. “We can, however, give a lower-cost version by giving up arm articulation, battery life, or communications range. The rapid increase in low-cost computing allows us to take an EOD tech in a developing country who would have to interact directly with an IED [improvised explosive device] and put a robot there instead.”
“There are technology challenges when you pass on high-cost components that do everything we want or lower-cost ones that do everything we need,” said Clayton. “It’s an interesting trade-off between, say, a low-cost motor with lower torque, less power for going on an incline. How much power do we need? Can we source motors from a company in the U.S. but assemble everything there?”
“Maybe we’ll get our body cast locally,” Clayton continued. “There are nice pockets of manufacturing capability in Cambodia, but maybe we’ll get our circuit board from somewhere else.” He described getting parts as a “next-level challenge — we have a good handle on where we’re going to source components and have manufacturing done, but it’s not yet on paper.”
The robot currently has a digital webcam and an outdoor pan-tilt-zoom camera, similar to those used in home surveillance. “The goal is to have three high-def cameras — one on a post to give a good view of the robot and its surroundings, one on the arm back a little bit from the claw, and one more on the front to inspect,” said Benson. “The point is to enable the operator not to be in line of sight of the robot, like in a building or dense vegetation.”
One hundred meters is considered a safe distance, and Benson noted that the controller would likely be in the form of a laptop in a ruggedized case.
Benson has been focusing on software and safety. “If there’s a lost connection, we want the robot to stop moving, or it could run over a landmine or harm the operator,” he said.
“Development included Arduino and Raspberry Pi,” said Clayton. “Again, good functionality but low cost. These components allow us to make the robot easy to use and repair in the field. Our control box is modular. We envision a troubleshooting tree that could suggest, for instance, that the field tech could pop in a spare control box and send the old one back to Golden West for repairs.”
As for reliability, the Low-Cost EOD Robot is designed to be “field-reparable,” Clayton said. “For example, we’ve given up some maneuverability in favor of robustness. Instead of using DC motors, we’ve gone with linear actuators that are water- and dustproof. The chances they go bad are reduced.”
The robot isn’t armored, because blast armor would raise the overall price. Villanova and Golden West want to make it available for under $10,000, a fraction of what military models cost.
Trials and the learning experience
The first two field tests further revealed the need for a strong supply chain and clear communications between the university team in Pennsylvania and people in Cambodia. These hurdles could trip up any long-distance international collaboration.
“In the winter of 2013-2014, we got our first taste of what it means to be in the field,” said Clayton. “It didn’t go so well.”
The team ran into heat issues. “We were there in their cool season, 98 degrees [Fahrenheit] with humidity, which caused a few problems for electronics,” he said. “And our component voltage regulator fried, and we couldn’t just run to a store or order online and expect speedy delivery.”
Clayton said that the team overcame the hurdle by hacking together a USB power bank and a USB car charger.
“We bought batteries in country, but their specs didn’t meet the required current draw of the motors, so we wrecked the batteries,” he said. “We ended up using car batteries that couldn’t fit in our battery compartment. The end goal is to have a locally sourced battery. This should be feasible since moped batteries should work and are available, given the number of mopeds in Cambodia.”
“Our collaborators at Golden West are quite tech-savvy, but they also take the role of customer,” said Clayton. “While our first claw prototype met all the stated specs, it wasn’t what they were looking for. Since one of the goals of this collaboration is to educate students, I’m glad we had this experience, which will make them better designers when they take the next step into industry.”
In January, the team ran more field tests in Cambodia. “This time, we took suitcases of extra parts, including batteries,” said Clayton. “We had some bugs, but Golden West was happy enough to say, ‘Let’s explore commercializing it.'”
Benson had been to Singapore before joining the team in Cambodia, and he learned the value of seeing where the robot will be used.
“Cambodia is more rural, still developing. I saw a lot of people missing limbs.” he said. “I realized from an engineering perspective that everything that comes naturally to us, like ordering an Arduino circuit board online, none of that exists. The need to design a robot in a different way stuck out.”
“In a place like Cambodia, many things are run out of small shops,” Clayton said. “If you know what you want, you can usually find someone who knows how to get it. Imagine finding someone who sells Arduino boards — we haven’t found him yet.”
The researchers plan to conduct another field test in June, and Clayton said he hopes that the robot is about a year to a year and a half to commercial delivery. “One of the concerns going forward is failure due to heat, but we have a lot of ideas on the table for cooling systems,” he added.
“We want users to be able to fix it in the field rather than have to stop a critical mission,” he said.
Getting ready for market
For the Villanova team and Golden West, the effort to complete the Low-Cost EOD Robot isn’t about making a profit but serving a need. “The market isn’t huge, so big companies haven’t looked at it yet,” said Clayton. “The partnership between the university and an NGO was a key enabler.”
The first year of research was self-funded by Villanova and Golden West. Since then, the project has gotten funding from an Engineering Education Partnership grant from the U.S. Department of State Office of Weapons Removal and Abatement, said Allen Tan, director of the design lab and Cambodia programs at Golden West.
The robot will be sold to organizations with EOD experience rather than farmers. “A potential customer is the Cambodian Mine Action Center,” said Clayton.
While the robot’s low cost might appeal to rescue or law enforcement agencies, Clayton and Tan noted that U.S. bomb squads are better funded than those in Asia or Africa and that the robot’s limited functionality probably wouldn’t interest them.
“We are in the early stages of commercialization, since the project started with an educational focus,” said Clayton. “Although we don’t have market projections; we’re excited to make this robot available in the developing world and hope it is well received.”