A patient recovering from a limb amputation won’t use a prostheses that isn’t comfortable. A person rehabbing from a stroke won’t use a robotic exoskeleton if the mobility it grants doesn't allow them to perform everyday activities. And a diabetes patient won’t use an insulin pump that doesn’t deliver the appropriate dosage of medicine to control their blood sugar.<img src="https://s3-us-west-1.amazonaws.com/wevolver-project-images/froala%2F1729896386001-IMG_8127_0_0.jpg" style="width: 100%px;" class="fr-fic fr-dib">Human-in-the-loop optimization algorithms could greatly improve human-robot interaction with medical devices like this exosuit, dramatically improving the results for patients. (Harvard Biodesign Lab)All of these challenges fall under a concept known as “human-robot interaction,” a critical issue in robotic design and engineering. A new perspective from researchers at the <a href="https://seas.harvard.edu/">Harvard John A. Paulson School of Engineering and Applied Sciences</a> (SEAS) delves into a potential approach to improve human-robot interaction. The approach, known as “human-in-the-loop optimization” (HILO), explores the potential of combining individual patient data with machine learning algorithms as a way to fine-tune robotic design and improve results for specific performance metrics. Patrick Slade, Assistant Professor of Bioengineering at SEAS, was first author on the perspective, which was published in Nature.“I hope this perspective can accelerate the development and effectiveness of human-robot interactions when developing robotic systems, especially for medical applications where these systems interact with people,” Slade said. “I’m hopeful that this perspective will encourage researchers to think about humans as complex, challenging systems to design for. HILO can be a tool to maximize the effectiveness of devices that interact with people.”The idea of human-in-the-loop optimization has been around for about 7 years, Slade said, but has yet to be fully adopted into the robotic engineering space. Because every human is different, human-robot interaction is a very difficult concept to predict and design for, which can slow down innovation in the field.“Working with these populations and devices for a while, I’ve seen that when people are introduced to a new device, their conceptions of the device, what they want and how they might use it, change as they get used to it,” he said. “It’s a moving target with dual challenges, where you need them to understand and adapt with the device, and you also want to provide them with something that provides a benefit to the user and is acceptable for regular use. HILO can help inform what the human-robot interaction should be to improve whatever human-robot metric is important to that population. Knowing what effective human-robot interaction is needed, for example, what torque to provide with an ankle exoskeleton to help people post-stroke to walk more quickly and efficiently, can inform the development of effective devices. HILO can also maximize performance of an existing commercial device by systematically tuning how that device interacts with the user. We’ve seen an optimized assistance double the benefits compared to a generic controller for an ankle exoskeleton device, showing HILO can have a big impact on patient outcomes without even changing the device hardware."<blockquote>One challenge in the field of robotics is that people often get treated as robots.Patrick Slade, Assistant Professor of Bioengineering at SEAS</blockquote>Though called an optimization algorithm, Slade described it as closer to a “personalization” algorithm, in that it uses personal data to optimize the device to the individual patient. And because it’s based on machine learning algorithms, the optimal adjustments can be found quickly and without an engineer needing to pour through multiple potential designs.“One challenge in the field of robotics is that people often get treated as robots,” Slade said. “People are complex and often don’t respond to robotic assistance in a predictable way. This paper is posing the idea that a lot of our research intuitions on human-robot interaction aren’t great because people are very complex. If we admit that, we can use this systematic approach to optimize assistance and really accelerate the field. Adopting HILO would really accelerate the field, and patients would see more benefits.”Slade envisions HILO as especially valuable in the earliest design stages, when research is still being done in labs as opposed to field testing finished or nearly finished designs. During those early stages, engineers typically test out numerous different designs and functions to try to hone in on one that meets whatever they determine to be the most important performance metrics. Examples of metrics in medical devices, for example, include muscle contraction,fatigue, walking speed, heart rate, blood glucose levels, or other measures of improved mobility.Using HILO algorithms, designers could quickly determine if a design needs to be adjusted based on individual patient data to maximize its benefit.“HILO allows you to identify the meaningful interactions, then build a portable device based on what you’ve learned,” he said. “Rather than building multiple interactions of a portable device, you could do it in a single shot, which could help accelerate the field.”Conversely, Slade’s perspective suggests even late-stage design or products already on the market could still benefit from these optimization algorithms. HILO is similar to getting a glasses prescription, systematically tuning the human-robot interaction to achieve a desired outcome, similar to a new prescription.“You have a device, you optimize the controller, and now you have a more personalized controller and see better benefits,” he said. “We’ve seen from a few studies that existing portable devices can double their performance if you apply HILO. It’s the same hardware – you just update the controller, and all of a sudden your device is twice as effective. That’s exciting. You could spend years developing something, and then all of a sudden double the performance just by systematically tuning the human-robot interaction.”

Human-in-the-loop optimization algorithms could greatly improve human-robot interaction with medical devices like this exosuit, dramatically improving the results for patients. (Harvard Biodesign Lab)

New research touts effectiveness of individual optimization algorithms

Keeping humans in the loop of robotic design

This article was discussed in our <a href="https://www.wevolver.com/profile/the.next.byte" rel="noopener noreferrer" target="_blank">Next Byte podcast</a>.&nbsp;<iframe height="200px" width="100%" frameborder="no" scrolling="no" seamless="" src="https://player.simplecast.com/3ddb6ef3-e0b1-4983-87ca-84cf2ed8fe2e?dark=false" class="fr-draggable" data-lf-form-tracking-inspected-kn9eq4roawkarlvp="true" data-lf-yt-playback-inspected-kn9eq4roawkarlvp="true" data-lf-vimeo-playback-inspected-kn9eq4roawkarlvp="true"></iframe> The full article will continue below.<hr>A set of knee exoskeletons, built with commercially available knee braces and drone motors at the University of Michigan, has been shown to help counteract fatigue in lifting and carrying tasks. They helped users maintain better lifting posture even when tired, a key factor in defending against on-the-job injuries, the researchers say.“Rather than directly bracing the back and giving up on proper lifting form, we strengthen the legs to maintain it,” said <a href="https://gregg.engin.umich.edu/" target="_blank" rel="noreferrer noopener">Robert Gregg</a>, U-M associate professor of robotics and corresponding author of the study in Science Robotics. “This differs from what’s more commonly done in industry.”Already workers who lift regularly, such as in construction and manufacturing, may use back braces. Back exoskeletons, which use springs or motors to help with lifting, are an emerging technology. But devices supporting the back assume unsafe lifting, or stooping, and back exoskeletons tend to be cumbersome devices that must be deactivated to allow motions that aren’t part of the lifting task, Gregg said.The Michigan team says their knee exoskeletons are the first to support the quadriceps muscles instead, which provide the bulk of the force in safe squat lifting, as a less intrusive way to help protect workers from back injuries. Study participants tested them out with lifting and carrying tasks using a 20 lb kettlebell.The tasks included lifting the weight off the ground and setting it down again, and lifting and carrying the weight on flat ground, up and down an incline, and up and down stairs. The study found that, after becoming fatigued, the participants kept better posture with the help of the exoskeleton, and they also lifted faster—just 1% slower than their pre-fatigued paces, versus 44% slower without the aid of the exoskeletons.&nbsp;“This is especially important when a worker has to keep up with a conveyor belt. Usually, when a worker is fatigued, they’ll keep up with that rate, but with a compromised posture. They’ll bend their back more, and that’s when injuries are most likely,” said Nikhil Divekar, a postdoctoral research fellow in robotics at U-M and first author of the study.&nbsp;The participants felt the benefit, too—they mainly said they were quite or very satisfied, with the exception of walking on level ground, for which they were more or less satisfied. This tracks with the small amount of assistance required by the quadriceps during this relatively easy task; Gregg described it as just enough support to counteract the weight of the exoskeleton.<img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNzI3MzkzMDMxNzk5LWtuZWUtZXhvcy12ZXJ0aWNhbC5qcGciLCJlZGl0cyI6eyJyZXNpemUiOnsid2lkdGgiOjk1MCwiZml0IjoiY292ZXIifX19" style="width: 100%px;" class="fr-fic fr-dib">Emily Keller, a doctoral student in robotics at the Locomotor Control Systems Lab, University of Michigan, demonstrates the knee exoskeletons on the steps outside of the Ford Robotics Building at U-M. Photo: José Montes-Pérez, Robotics, University of MichiganOne of the keys to making the exoskeleton so wearable is the motors and the way that they are geared, which enables users to swing their knees freely for a natural gait. The other is the software, which predicts what kind of assistance the user needs by measuring the angle of the knee joint, the orientations of the thigh and lower leg, and the force picked up by a sensor in the user’s shoe.&nbsp;With these three measurements from both legs, it’s possible to work out what motion the user is trying to do, and how much assistance to give. These measurements were taken 150 times per second, enabling the exos to move seamlessly between activities.&nbsp;This approach is in contrast to many exo controllers, which follow predefined patterns for a limited set of tasks. Switching tasks can be a problem for these controllers, and they may need a full second to figure out what the user is trying to do, Gregg said.“If your exo is trying to walk upstairs, and you’re trying to walk downstairs, that could be a problem, right?” he said.The new controller also pairs a physics model with machine learning, which prevents the exoskeleton from making unexpected moves if the user begins behaving differently than any activity included in the controller’s training data.The lab prototypes cost about $4,000 per pair, so Gregg anticipates that if the exoskeletons were produced at scale, they might cost about $2,000 per pair.The 10 study participants, five women and five men, did all the tasks on two different days, one day fresh and the other fatigued. To induce fatigue, each participant performed a series of squat lifts with the kettlebell until they couldn’t continue without a long break between repetitions. All participants had experience with proper squat lifting techniques.&nbsp;The study was funded by the National Institutes of Health.The team has applied for patent protection with the assistance of U-M Innovation Partnerships and is seeking partners to bring the technology to market.Gray Thomas, Avani Yerva and Hannah Frame contributed to the study.Gregg is also an associate professor of electrical and computer engineering and mechanical engineering.

Helping out the quad muscles kept study participants lifting safely despite fatigue, with an algorithm that smoothly shifts between lifting and carrying tasks.

Versatile knee exo for safer lifting

Cloud Ground Control

APEX Motion Control-Fri Jun 30 2023 23:59:59 GMT+0200 (Central European Summer Time)

APEX Motion Control-Mon Jul 31 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

APEX Motion Control-Thu Aug 31 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

APEX Motion Control-Thu Aug 31 2023 23:59:59 GMT+0200 (Central European Summer Time)

APEX Motion Control-Sat Sep 30 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

Cloud Ground Control-Fri Jun 30 2023 23:59:59 GMT+0200 (Central European Summer Time)

Cloud Ground Control-Mon Jul 31 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

Cloud Ground Control-Thu Aug 31 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

Cloud Ground Control-Thu Aug 31 2023 23:59:59 GMT+0200 (Central European Summer Time)

Cloud Ground Control-Sat Sep 30 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

The Next Byte-Fri Jun 30 2023 23:59:59 GMT+0200 (Central European Summer Time)

The Next Byte-Mon Jul 31 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

The Next Byte-Thu Aug 31 2023 23:59:59 GMT+0200 (Central European Summer Time)

The Next Byte-Thu Aug 31 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

The Next Byte-Sat Sep 30 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

The Things Network-Fri Jun 30 2023 23:59:59 GMT+0200 (Central European Summer Time)

The Things Network-Mon Jul 31 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

The Things Network-Thu Aug 31 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

The Things Network-Thu Aug 31 2023 23:59:59 GMT+0200 (Central European Summer Time)

The Things Network-Sat Sep 30 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

The Linley Group-Fri Jun 30 2023 23:59:59 GMT+0200 (Central European Summer Time)

The Linley Group-Mon Jul 31 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

The Linley Group-Thu Aug 31 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

The Linley Group-Thu Aug 31 2023 23:59:59 GMT+0200 (Central European Summer Time)

The Linley Group-Sat Sep 30 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

The National Robotarium-Fri Jun 30 2023 23:59:59 GMT+0200 (Central European Summer Time)

The National Robotarium-Mon Jul 31 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

The National Robotarium-Thu Aug 31 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

The National Robotarium-Thu Aug 31 2023 23:59:59 GMT+0200 (Central European Summer Time)

The National Robotarium-Sat Sep 30 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

3D Printing in Medicine-Fri Jun 30 2023 23:59:59 GMT+0200 (Central European Summer Time)

3D Printing in Medicine-Mon Jul 31 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

3D Printing in Medicine-Thu Aug 31 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

3D Printing in Medicine-Thu Aug 31 2023 23:59:59 GMT+0200 (Central European Summer Time)

3D Printing in Medicine-Sat Sep 30 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

FACTUREE – The Online Manufacturer-Fri Jun 30 2023 23:59:59 GMT+0200 (Central European Summer Time)

FACTUREE – The Online Manufacturer-Mon Jul 31 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

FACTUREE – The Online Manufacturer-Thu Aug 31 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

FACTUREE – The Online Manufacturer-Thu Aug 31 2023 23:59:59 GMT+0200 (Central European Summer Time)

FACTUREE – The Online Manufacturer-Sat Sep 30 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

human in the loop control

Human-In-The-Loop Control

Profiles