Seated Postural and Reaching Control Training with the Trunk Support Trainer (TruST)
“We wanted to scientifically demonstrate how robotic TruST can be used to deliver an intense activity-based postural and reaching training to improve the functional sitting abilities of children with CP and trunk control problems”, says Victor Santamaria, a physical therapist and associate researcher scientist in Agrawal’s Robotics and Rehabilitation Laboratory, and first author of the paper.
Recent developments in robotic equipment have enabled clinicians to address engagement, repetition, and intensity for their patients to practice task-oriented movements in CP. A team led by Agrawal, together with other researchers at Teacher’s College and the Columbia University Irving Medical Center, recently won a five-year National Institutes of Health R01 award (#1R01 HD101903-01) to conduct a randomized clinical trial.
The project—"Improving seated postural control and upper extremity function in bilateral CP with a robotic Trunk-Support-Trainer (TruST)"—will involve up to 80 children with poor trunk control. Some will use the TruST robotic rehabilitation while others will try conventional rehabilitation. This new NIH study will compare the efficacy of the motorized TruST to engage children in play-oriented practice while advancing their skill progression with static trunk support.
“Our new NIH project is a randomized clinical trial with a large sample size to study the efficacy of TruST-intervention as a unique therapeutic solution to promote seated functional abilities in children with bilateral CP,” Agrawal adds.
About the Study
Journal: IEEE Transactions of Neural Systems and Rehabilitation Engineering
Title: Promoting Functional and Independent Sitting in Children with Cerebral Palsy Using the Robotic Trunk Support Trainer
Authors: Victor Santamaria, Moiz Khan, Tatiana Luna, Jiyeon Kang, Joseph Dutkowsky, Andrew Gordon, and Sunil Agrawal, Department of Mechanical Engineering, Columbia Engineering.
The pilot study was partially funded by the Langer Foundation as administered by The Order of Malta.
COI: The authors declare no financial or other conflicts of interest.
Columbia Engineering
Columbia Engineering, based in New York City, is one of the top engineering schools in the U.S. and one of the oldest in the nation. Also known as The Fu Foundation School of Engineering and Applied Science, the School expands knowledge and advances technology through the pioneering research of its more than 220 faculty, while educating undergraduate and graduate students in a collaborative environment to become leaders informed by a firm foundation in engineering. The School’s faculty are at the center of the University’s cross-disciplinary research, contributing to the Data Science Institute, Earth Institute, Zuckerman Mind Brain Behavior Institute, Precision Medicine Initiative, and the Columbia Nano Initiative. Guided by its strategic vision, “Columbia Engineering for Humanity,” the School aims to translate ideas into innovations that foster a sustainable, healthy, secure, connected, and creative humanity.
Header image: Trunk Support Trainer (TruST).
Human-Robot Facial Co-expression
Go inside the Creative Machines Lab to watch Emo's facial co-expression. Video Credit: Yuhang Hu, Columbia Engineering
What’s next
The researchers are now working to integrate verbal communication, using a large language model like ChatGPT into Emo. As robots become more capable of behaving like humans, Lipson is well aware of the ethical considerations associated with this new technology.
“Although this capability heralds a plethora of positive applications, ranging from home assistants to educational aids, it is incumbent upon developers and users to exercise prudence and ethical considerations,” says Lipson, James and Sally Scapa Professor of Innovation in the Department of Mechanical Engineering at Columbia Engineering, co-director of the Makerspace at Columbia, and a member of the Data Science Institute. “But it’s also very exciting -- by advancing robots that can interpret and mimic human expressions accurately, we're moving closer to a future where robots can seamlessly integrate into our daily lives, offering companionship, assistance, and even empathy. Imagine a world where interacting with a robot feels as natural and comfortable as talking to a friend.”
About the Study
Journal: Science Robotics
Title: “Human-Robot Facial Co-expression”
Authors: Yuhang Hu (1); Boyuan Chen (2, 3, 4); Jiong Lin (1); Yunzhe Wang (5); Yingke Wang (5); Cameron Mehlman (1); and Hod Lipson (1, 6)
- Creative Machines Laboratory, Mechanical Engineering Department, Columbia University,
- Mechanical Engineering and Materials Department, Duke University
- Electrical and Computer Engineering Department, Duke University
- Computer Science Department, Duke University
- Computer Science Department, Columbia University
- Data Science Institute, Columbia University
Funding: The study was supported by the National Science Foundation AI Institute for Dynamical Systems (DynamicsAI.org ) grant 2112085, and an Amazon grant through the Columbia Center of AI Technology (CAIT).
COI: The authors declare that they have no competing interests.
Daniela Durón García
Mary Boyce
Melbourne Francis
Club Life: Columbia University Formula SAE
Build an electric race car from scratch? Go behind the scenes to see Columbia’s Formula Racing club assemble the key components of their racing vehicle.
Safety first
Back in the basement workshop, the club’s president, Sophia Ladyzhets BS’22, and Catherine “Calee” Schmidtberger BS’22, the club’s chief mechanical engineer, stand surrounded by car parts, from v-shaped suspension rods to the braking system to the hub and uprights assembly holding these components and the tires. The team has already run dozens of tests on each system, checking and double checking that all fail safes are in place and functional. The group has even created a full-sized model of the car’s front end to study how well it can withstand heavy impact.
Every aspect of the design process needs to be timed and executed with precision. Ben Felson BS’24, the team’s vice president of technical operations, pulls out an Excel spreadsheet containing a long list containing hundreds upon hundreds of FSAE rules about vehicle systems, subsystems, and subsystems for those subsystems.
“I have these little rules readings on the weekends where we literally go through and read each one, talk about it and what it’s about, then establish who is in charge of knowing it and when we reviewed it,” Felson said. Because the motor, high voltage battery pack, and other integral electrical components sit right behind the driver, “the rules are more high stakes for the EV because parts of the system can be a real danger to the driver if handled incorrectly.”
But the beating heart of any car is the motor, and, in an electric vehicle, that’s where things get particularly interesting. Hovering over the open accumulator, Partida described the internal functions of the car’s powertrain, from the battery pack, the relays and electronics to the motor controller and the electric motors, with surgeon-like detail. The battery pack is a layer cake of power, with each battery segment containing multiple clusters of cells. Members hope that those cells will take their car upwards of speeds of 120 kilometers per hour.
That’s not just about raw power; it’s also about engineering an incredibly effective use of it.
“The efficiency of the motor inverter [the EV component that converts direct current to alternating current] is upwards of 90%,” Schmidtberger said. “So the power efficiency is very high, especially compared to internal combustion cars, where the standard efficiency of an engine is around 35%.”
Indeed, it’s never as simple as taking the bones of a combustion car and swapping in a few electric parts.
“Our first iteration of this prototype used an old chassis, the old bones that were from an old internal combustion car,” Schmidtberger said. “In that case, several of our parts were wonky and ended up having a horrible shape that would have never worked. Our new chassis is designed to be an electric car that can fit the accumulator in the back to maintain a low center of gravity, preserving its handling capabilities.”
A new generation of car designers
The push toward electric design had the side benefit of attracting a wide range of students, not just the stereotypical motorhead crowd of yesteryear, noted Professor of Mechanical Engineering Jeffrey Kysar, another of the club’s faculty advisors. “The refocus is attracting students not just from mechanical and electrical engineering, but from chemical, civil, and earth and environmental engineering—as well as from other schools and laboratories at Columbia such as the Lamont-Doherty Earth Observatory,” he said. (Test drives are conducted in the parking lot of Lamont-Doherty, thanks to its director Maureen Raymo.)
“Sustainability is a global problem that attracts everyone,” said Preindl.
Ladyzhets noted there have been several non-engineering students joining the club, such as Elaine Kharbanda ’24, the club’s vice president of business affairs and a linguistics student at Columbia College. At the competition, the team’s inclusivity efforts were recognized by General Motors’ “Everybody In” award, which is presented to the team that best reflected the company’s Everybody In campaign. As part of their prize, GM will cover the group’s registration fees for next year’s competition.
Browne added that the team’s diversity is starting to be reflected within the wider industry, citing GM’s own Mary Barra, who was named CEO in 2014. He also acknowledged Mary Boyce, now provost of the University, who ramped up funding for the club and also completely renovated the FSAE’s workshop back when she was dean of the engineering school. Browne said it was her interest in FSAE that made the transition to EV possible.
Transitions are much on the minds of club members, as summer marks the end of another academic year. Ladzyzhets is pursuing her integrated master’s degree in mechanical engineering at Columbia, and is planning for an upcoming internship with Tesla. Schmidtberger, on the other hand, wants to follow her childhood dream and build energy-efficient amusement park rides, or also get involved in other sustainable projects.
But most importantly, the legacy they want to leave behind is one of inclusivity, where everyone can feel like they’re building the future together.
“With the electric car, we’ve tried to make it a lot more welcoming—there’s no experience needed—we don’t do interviews, and we’re trying to put a lot more work into our teaching and mentorship,” Ladyzhets said. “We’re all trying to figure out these problems together, and that’s the most important thing.”
Highly dexterous robot hand even works in the dark
Researchers at Columbia Engineering have demonstrated a highly dexterous robot hand, one that combines an advanced sense of touch with motor learning algorithms in order to achieve a high level of dexterity.
As a demonstration of skill, the team chose a difficult manipulation task: executing an arbitrarily large rotation of an unevenly shaped grasped object in hand while always maintaining the object in a stable, secure hold. This is a very difficult task because it requires constant repositioning of a subset of fingers, while the other fingers have to keep the object stable. Not only was the hand able to perform this task, but it also did it without any visual feedback whatsoever, based solely on touch sensing.
In addition to the new levels of dexterity, the hand worked without any external cameras, so it's immune to lighting, occlusion, or similar issues. And the fact that the hand does not rely on vision to manipulate objects means that it can do so in very difficult lighting conditions that would confuse vision-based algorithms--it can even operate in the dark.
“While our demonstration was on a proof-of-concept task, meant to illustrate the capabilities of the hand, we believe that this level of dexterity will open up entirely new applications for robotic manipulation in the real world,” said Matei Ciocarlie, associate professor in the Departments of Mechanical Engineering and Computer Science. “Some of the more immediate uses might be in logistics and material handling, helping ease up supply chain problems like the ones that have plagued our economy in recent years, and in advanced manufacturing and assembly in factories.”
Leveraging optics-based tactile fingers
In earlier work, Ciocarlie’s group collaborated with Ioannis Kymissis, professor of electrical engineering, to develop a new generation of optics-based tactile robot fingers. These were the first robot fingers to achieve contact localization with sub-millimeter precision while providing complete coverage of a complex multi-curved surface. In addition, the compact packaging and low wire count of the fingers allowed for easy integration into complete robot hands.
Teaching the hand to perform complex tasks
For this new work, led by CIocarlie’s doctoral researcher, Gagan Khandate, the researchers designed and built a robot hand with five fingers and 15 independently actuated joints--each finger was equipped with the team’s touch-sensing technology. The next step was to test the ability of the tactile hand to perform complex manipulation tasks. To do this, they used new methods for motor learning, or the ability of a robot to learn new physical tasks via practice. In particular, they used a method called deep reinforcement learning, augmented with new algorithms that they developed for effective exploration of possible motor strategies.
Robot completed approximately one year of practice in only hours of real-time
The input to the motor learning algorithms consisted exclusively of the team’s tactile and proprioceptive data, without any vision. Using simulation as a training ground, the robot completed approximately one year of practice in only hours of real-time, thanks to modern physics simulators and highly parallel processors. The researchers then transferred this manipulation skill trained in simulation to the real robot hand, which was able to achieve the level of dexterity the team was hoping for. Ciocarlie noted that “the directional goal for the field remains assistive robotics in the home, the ultimate proving ground for real dexterity. In this study, we've shown that robot hands can also be highly dexterous based on touch sensing alone. Once we also add visual feedback into the mix along with touch, we hope to be able to achieve even more dexterity, and one day start approaching the replication of the human hand.”
Ultimate goal: joining abstract intelligence with embodied intelligence
Ultimately, Ciocarlie observed, a physical robot being useful in the real world needs both abstract, semantic intelligence (to understand conceptually how the world works), and embodied intelligence (the skill to physically interact with the world). Large language models such as OpenAI’s GPT-4 or Google’s PALM aim to provide the former, while dexterity in manipulation as achieved in this study represents complementary advances in the latter.
For instance, when asked how to make a sandwich, ChatGPT will type out a step-by-step plan in response, but it takes a dexterous robot to take that plan and actually make the sandwich. In the same way, researchers hope that physically skilled robots will be able to take semantic intelligence out of the purely virtual world of the Internet, and put it to good use on real-world physical tasks, perhaps even in our homes.
The paper has been accepted for publication at the Robotics: Science and Systems Conference (Daegu, Korea, July 10-14, 2023), and is currently available as a preprint.
ABOUT THE STUDY
CONFERENCE: Science and Systems Conference (Daegu, Korea, July 10-14, 2023)
STUDY: "Sampling-based Exploration for Reinforcement Learning of Dexterous Manipulation”
AUTHORS: Authors are all from Columbia Engineering: Gagan Khandate and Tristan Luca Saidi (Computer Science), Siqi Shang, Eric Chang, Johnson Adams, and Matei Ciocarlie (Mechanical Engineering). The tactile sensors were developed in collaboration with Ioannis Kymissis (Electrical Engineering).
FUNDING: This work was supported in part by the Office of Naval Research grant N00014-21-1-4010 and the National Science Foundation grant CMMI-2037101.
The authors declare no financial or other conflicts of interest.