<p>The drone, or Unmanned Aerial Vehicle (UAV), has transformed the practice of land surveying, providing a faster, more efficient way to complete large-area surveys complemented by high-resolution visual and thermal aerial images.</p><p>Unfortunately, the two types of conventional drone in common use for surveying today – quadcopters, and fixed-wing craft – both have drawbacks which reduce their effectiveness in surveying and mapping applications.</p><p>Now an ambitious and innovative drone manufacturer, <a href="https://www.atmosuav.com/" rel="noopener" target="_blank">ATMOS UAV</a> based in Delft, The Netherlands, is disrupting the drone market with a breakthrough design combining take-off and landing from helicopters with the long endurance of airplanes. ATMOS UAV’s goal is to provide the ultimate airborne surveying tool offering the industry’s best all-round performance. To achieve the millimeter-level accuracy that ATMOS UAV’s customers value, it relies on the <a href="https://www.xsens.com/mti-products" rel=" noopener" target="_blank">MTi series</a> of integrated <a href="https://www.xsens.com/inertial-sensor-modules" rel=" noopener" target="_blank">inertial sensor modules</a> from Xsens.</p><p><strong>Easy to launch, fly and land</strong></p><p>Of the two conventional types of drone, the fixed-wing type is better for surveying missions once it is in the air: it cruises faster and more efficiently than the other conventional type of drone, the quadcopter. This means that it can sustain longer flights and cover a greater survey area on each mission. It also offers greater stability in flight and is less affected by wind – an important characteristic for surveying missions which need to capture sharp, high-resolution visual and thermal images, unaffected by blur or other visual artefacts.</p><p>The biggest advantage is its manoeuvrability, which leads to the ability to take-off and land from a small footprint using&nbsp;its Vertical Take-Off and Landing (VTOL) capability. Unlike a fixed-wing craft, it requires no runway or slingshot launching device, and operates from a tiny landing area.</p><p>The breakthrough made by ATMOS was to develop Marlyn, a hybrid fixed-wing UAV which performs vertical take-off and landing from a 2m x 2m landing area, but which cruises smoothly and efficiently.</p><p><span class="fr-video fr-deletable fr-fvc fr-dvb fr-draggable" contenteditable="false" draggable="true"><iframe width="640" height="360" src="https://www.youtube.com/embed/LtcLD3zOHAc?&amp;wmode=opaque&amp;enablejsapi=1&amp;origin=https://www.wevolver.com" frameborder="0" allowfullscreen="" class="fr-draggable" data-gtm-yt-inspected-37="true" data-lf-form-tracking-inspected-kn9eq4roawkarlvp="true" data-lf-yt-playback-inspected-kn9eq4roawkarlvp="true" data-lf-vimeo-playback-inspected-kn9eq4roawkarlvp="true"></iframe></span><br></p><p>&nbsp;The fully-electric Marlyn drone includes a redundant dual battery system which enables it to fly uninterrupted for 50 minutes, reaching a top speed of 95km/h when cruising. The flight plan is easy to program before each flight in ATMOS’ intuitive Marlynk software tool. The drone offers impressive stability in flight: it can handle wind speeds up to 6 Bft (45km/h) during take-off, cruising and landing.</p><div class="u_fNK" data-grammarly-part="highlights" style="position: absolute; top: 0px; left: 0px;"><div style="box-sizing: content-box; top: 483px; left: 0px; width: 778px; height: 39px; position: relative; pointer-events: none; overflow: hidden; border: 0px; border-radius: 0px; padding: 0px; margin: 0px;"><div style="position: absolute; top: 0px; left: 0px;"><div style="height: 959px; width: 2485px;"><div><br></div><div><br></div></div></div></div></div><div class="u_fNK" data-grammarly-part="button" style="position: absolute; top: 0px; left: 0px;"><div style="box-sizing: content-box; top: 483px; left: 0px; width: 778px; height: 39px; position: relative; pointer-events: none; overflow: hidden; border: 0px; border-radius: 0px; padding: 0px; margin: 0px;"><div style="position: absolute; transform: translate(-100%, -100%); top: 29px; left: 773px; width: auto; height: auto; pointer-events: all;"><div style="flex-direction: row; display: none; pointer-events: auto;"><div style="position: relative;"><div class="_2vV1Z _3uWIx" style="position: absolute; right: 0px; top: 0px;"><div class="_2mmtW"><div class="-mjUa"><div class="_1NNY0 _3PC-V" style="transform: scale(1);">0</div>&nbsp;Advanced issue found<div class="SY83w _5pAkg">▲</div><div class="_3pmCo"><br></div></div></div></div></div><div><div class="_5WizN _1QzSN"><div class="_3YmQx"><div class="_3QdKe" title="Protected by Grammarly">&nbsp;</div></div></div></div></div></div></div></div><p><br></p><div class="u_fNK" data-grammarly-part="highlights" style="position: absolute; top: 0px; left: 0px;"><div style="box-sizing: content-box; top: 483px; left: 0px; width: 778px; height: 39px; position: relative; pointer-events: none; overflow: hidden; border: 0px; border-radius: 0px; padding: 0px; margin: 0px;"><div style="position: absolute; top: 0px; left: 0px;"><div style="height: 959px; width: 2485px;"><div><br></div><div><br></div></div></div></div></div><div class="u_fNK" data-grammarly-part="button" style="position: absolute; top: 0px; left: 0px;"><div style="box-sizing: content-box; top: 483px; left: 0px; width: 778px; height: 39px; position: relative; pointer-events: none; overflow: hidden; border: 0px; border-radius: 0px; padding: 0px; margin: 0px;"><div style="position: absolute; transform: translate(-100%, -100%); top: 29px; left: 773px; width: auto; height: auto; pointer-events: all;"><div style="flex-direction: row; display: none; pointer-events: auto;"><div style="position: relative;"><div class="_2vV1Z _3uWIx" style="position: absolute; right: 0px; top: 0px;"><div class="_2mmtW"><div class="-mjUa"><div class="_1NNY0 _3PC-V" style="transform: scale(1);">0</div>&nbsp;Advanced issue found<div class="SY83w _5pAkg">▲</div><div class="_3pmCo"><br></div></div></div></div></div><div><div class="_5WizN _1QzSN"><div class="_3YmQx"><div class="_3QdKe" title="Protected by Grammarly">&nbsp;</div></div></div></div></div></div></div></div><p><span class="fr-img-wrap"><img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNTg4MzQwMjQwOTQzLU1hcmx5bmsuanBnIiwiZWRpdHMiOnsicmVzaXplIjp7IndpZHRoIjo5NTAsImZpdCI6ImNvdmVyIn19fQ==" alt="Marlynk2" srcset="https://www.xsens.com/hs-fs/hubfs/Marlynk2.png?width=408&amp;name=Marlynk2.png 408w, https://www.xsens.com/hs-fs/hubfs/Marlynk2.png?width=815&amp;name=Marlynk2.png 815w, https://www.xsens.com/hs-fs/hubfs/Marlynk2.png?width=1223&amp;name=Marlynk2.png 1223w, https://www.xsens.com/hs-fs/hubfs/Marlynk2.png?width=1630&amp;name=Marlynk2.png 1630w, https://www.xsens.com/hs-fs/hubfs/Marlynk2.png?width=2038&amp;name=Marlynk2.png 2038w, https://www.xsens.com/hs-fs/hubfs/Marlynk2.png?width=2445&amp;name=Marlynk2.png 2445w" sizes="(max-width: 815px) 100vw, 815px" class="fr-fic fr-dii"><span class="fr-inner" contenteditable="true" spellcheck="false"><em>The intuitive user interface of the Marlynk flight planning software</em>&nbsp;</span></span></p><p><br></p><p>ATMOS supplies Marlyn with a choice of surveying cameras which produce visual, thermal or multi-spectral outputs in all the common data formats supported by professional surveying software systems.</p><p><strong>Millimeter-scale geolocation accuracy</strong></p><p>The dynamic characteristics of Marlyn, then, enable surveyors to fly long missions over large survey areas, with the convenience and ease of use that VTOL capability offers.</p><p>But the real test of surveying equipment is its geolocational accuracy – and here Marlyn excels, thanks to its high-performance <a href="https://www.xsens.com/products/mti-100-series" rel=" noopener" target="_blank">MTi-G-710-GNSS</a> motion sensor module from Xsens. The MTi-G-710-GNSS is Xsens’ most full-featured sensor: it integrates a 3D accelerometer, 3D gyroscope, magnetometer and a receiver input for a GNSS satellite positioning signal into a small 37mm x 33mm x 12mm module which weighs just 11g. Low power consumption of just 750mW in normal operation helps ATMOS to prolong the Marlyn’s flight duration between battery charges.</p><p>The MTi-G-710-GNSS provides high-quality attitude, position, velocity, acceleration, magnetic heading and rate-of-turn data. The GNSS receiver features a multiple-constellation configuration which maintains high location accuracy, even during periods of limited GNSS visibility. Proprietary sensor fusion software developed by Xsens combines the unit’s inertial measurements with the satellite signals to produce accurate positioning data at an output data rate of up to 400Hz. In the Marlyn system, the survey outputs and visual/thermal images are tagged with geolocational data accurate to as little as ±10mm.</p><p>The sensor module’s attitude and heading data outputs are also used to stabilize the Marlyn drone in flight, and to enable optical image stabilization. The low latency of the Xsens sensor – less than 2ms – means that these stabilization systems can respond in real-time to events such as gusts of wind which disturb the drone’s attitude.</p><p>Despite the internal complexity of the MTi-G-710-GNSS module’s components, the outputs were easy for Marlyn’s developers to integrate into their system software, thanks to the MT Software Suite developer tool, supplied free by Xsens with any MTi module.</p><p><img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNTg4MzQwMzU0Njg4LUFUTU9TLVVBVi5qcGciLCJlZGl0cyI6eyJyZXNpemUiOnsid2lkdGgiOjk1MCwiZml0IjoiY292ZXIifX19" style="width: 800px;" class="fr-fic fr-dib"></p><p>Rene Worms, Global Head of Sales at ATMOS, says that the positional accuracy of Marlyn is the key to its position as the industry’s premium airborne surveying tool. He says: ‘Marlyn offers surveying professionals the highest productivity as well as the most accurate and detailed outputs. To be able to make this promise, we specify premium components which provide the best performance. That’s why we chose the MTi-G-710-GNSS, which gives us a unique combination of high accuracy, reliability, small size and weight, and ease of use.’</p><p><em>For more information about the Marlyn surveying drone, go to&nbsp;</em><a href="http://www.atmosuav.com/" target="_blank"><em>www.atmosuav.com</em></a><em>.</em></p>

Millimeter-level Accuracy Thanks To Xsens Motion Sensor

Exciting New Drone Design From ATMOS UAV

<p>Robots are gradually permeating into more and more areas of our daily life: smart cars are at work on storage facilities and production sites, they test delivery services. Researchers in St. Petersburg are designing a mobile autorobotic inspector (MARTI) that will help the State Road Traffic Safety Inspection officers by documenting details of traffic road accidents and detecting traffic code violations. ITMO.NEWS got in touch with Professor Oleg Basov, deputy dean of ITMO&rsquo;s Faculty of Infocommunication Technologies, to learn why one quadcopter can create a better accident plan than three inspectors, and how MARTI will be able to model traffic road accidents.&nbsp;</p><p>Can a quadcopter actually take on some of the duties of a State Road Traffic Safety inspector? According to researchers from ITMO University and Saint Petersburg State University of Architecture and Civil Engineering, it definitely can. Moreover, MARTI will be more efficient at documenting the evidence of an accident, creating its site layout plan, and even reconstructing the scene of the accident. Today, the researchers are already testing a prototype of the system.&nbsp;</p><p>It all started with research conducted by Prof. Basov&rsquo;s group at the Faculty of Infocommunication Technologies. For a while now, scientists and engineers have been trying to enhance the capabilities of standard quadcopters for professional use. They are developing special transport modules, almost like portable airfields for drones that could fit into a garage box on the roof of a car, or in its luggage compartment.&nbsp;</p><p>These devices will make it possible to transport quadcopters, launch and land them, as well as change their battery automatically, without an operator. The data received from the drone will also be transmitted here. All of that will not require any substantial changes to the vehicle. Thus, the changes will extend the use of quadcopters from the current 30-40 minutes to 3-4 hours, with the vehicle transporting it to the site and providing the necessary power supply. At one point, this project attracted the attention of a research group working on traffic road safety at Saint Petersburg State University of Architecture and Civil Engineering.</p><p><span class="fr-img-caption fr-fic fr-dib" style="width: 788px;"><span class="fr-img-wrap"><img src="https://s3-us-west-1.amazonaws.com/wevolver-project-images/froala%2F1586356143773-931207.jpg"><span class="fr-inner">Traffic road accident&nbsp;</span></span></span></p><blockquote><p>&ldquo;Our idea sounded interesting to Professor Sergey Evtukov, head of Road Safety Institute at Saint Petersburg State University of Architecture and Civil Engineering,&rdquo; says Prof. Basov. &ldquo;He suggested we use quadcopters to analyze traffic road accidents. Why is it important? Traffic safety inspectors can&rsquo;t always make it in time to the site of an accident, especially if it&rsquo;s caused a traffic jam. While they are on the way, some evidence can get lost or changed by the drivers to hide what has actually happened. A quadcopter, on the other hand, can launch straight away to capture and store the details of the accident. Drones can be transported on the inspectors&rsquo; car and fly ahead as they are getting to the site, or even have a base near an accident-prone area to launch at the alarm about a new accident. When the inspectors arrive at the site, information from the quadcopter will be transferred to their car and processed, allowing the inspectors to have a complete site layout plan at the moment of their arrival.&rdquo;</p></blockquote><p>Based on this information, inspectors will be able to make objective conclusions as to who caused the accident, which would be important when drivers contact their insurance companies, and in settling any legal arguments. MARTI will not only be able to determine the boundaries of the accident site and capture it, but also document the location of cars, and complexity of their mechanical damage, any marks on the roadway, or individual debris and objects on the site, all of which will later make it possible to visualize the accident in 3D.</p><blockquote><p>&ldquo;It will estimate the kind of incoming data &ndash; the speed, direction, maneuvers of cars and weather conditions &ndash; that could have caused the accident, and the kind that couldn&rsquo;t. Thus, it will allow us to determine the culprit of the accident,&rdquo; explains Prof. Basov.&nbsp;</p></blockquote><p><span class="fr-img-caption fr-fic fr-dib" style="width: 788px;"><span class="fr-img-wrap"><img src="https://s3-us-west-1.amazonaws.com/wevolver-project-images/froala%2F1586356222255-931204.jpg"><span class="fr-inner">The experiment at the Oryol Law Institute of the Ministry of Internal Affairs test range&nbsp;</span></span></span></p><h2><strong>One quadcopter against three inspectors</strong></h2><p>An experiment was conducted to test the capabilities of the drones and software developed at ITMO. A multifunctional test range at Oryol Law Institute of the Ministry of Internal Affairs of the Russian Federation named after V. V. Lukyanov was chosen as the site of the experiment.&nbsp;</p><blockquote><p>&ldquo;We went to the test range with a model of a city, placed two cars at the crossroads as if they bumped into each other, and then conducted the following experiment. The quadcopter was able to document the site, measure the distance between the cars, and create a model of the accident in about three minutes. Next came the inspectors &ndash; three people measured the distance with a measuring tape and drew a layout plan in ten minutes. Usually, it takes even longer, because there&rsquo;s only one inspector who documents everything with the help of witnesses,&rdquo; remembers Prof. Basov.&nbsp;</p></blockquote><p><span class="fr-img-caption fr-fic fr-dib" style="width: 788px;"><span class="fr-img-wrap"><img src="https://s3-us-west-1.amazonaws.com/wevolver-project-images/froala%2F1586356254077-931200.jpg"><span class="fr-inner">A traffic road accident model. Image provided by Prof. Basov&nbsp;</span></span></span></p><p>Moreover, last month, Oryol hosted a round-table &ldquo;Research of the application of unmanned flying vehicles to acquire primary information about traffic road accidents&rdquo; with representatives from ITMO University, Saint Petersburg State University of Architecture and Civil Engineering, Oryol Law Institute of the Ministry of Internal Affairs of the Russian Federation, Russian Federation Security Guard Service Federal Academy, St. Petersburg Institute for Informatics and Automation of the Russian Academy of Sciences, the Forensic Center at the Regional Office of the Ministry of Internal Affairs of Oryol Region, as well as traffic safety inspectors. The participants discussed how quadcopters can be used to help police officers on accident sites. Field tests of the system will also continue, as the first experiment has highlighted some software errors that are now being fixed by the researchers.&nbsp;</p><blockquote><p>&ldquo;We&rsquo;ve compared the results acquired with the drone to those from our inspectors,&rdquo; expounds Prof. Basov. &ldquo;There are some bugs, and we&rsquo;re now working to account for them. My colleagues are busy at the Oryol testing range looking for the perfect height and camera angles that would allow us to avoid errors. The next stage is designing a hardware and software complex to reenact the accident itself that would model the movement of cars.&rdquo;</p></blockquote><p><span class="fr-img-caption fr-fic fr-dib" style="width: 788px;"><span class="fr-img-wrap"><img src="https://s3-us-west-1.amazonaws.com/wevolver-project-images/froala%2F1586356279664-931201.jpg"><span class="fr-inner">A traffic road accident model with road markings. Image provided by Prof. Basov&nbsp;</span></span></span></p><h2><strong>A monorail for MARTI</strong></h2><p>These days, MARTI is already a finished concept and the researchers are acquiring a patent on it. Next comes its production.</p><blockquote><p>&ldquo;It will be a monorail placed on separating strips of highways, on bridges and black spots on the roads,&rdquo; continues Prof. Basov. &ldquo;MARTI will automatically move to the place where a quadcopter is needed. The monorail will allow it to be quick and effective regardless of traffic jams or any other obstacles.&rdquo;</p></blockquote><p>The research group is also developing algorithms for automatic flight of quadcopters. Thus, the drones will not only be able to capture the details of an accident, but also monitor speeding incidents, violations of between-car range and road marking. Such an algorithm, however, will require a more precise fine tuning of flight and positioning at a certain altitude, taking into account the wires, street lights and other obstacles.</p><p>Transport modules that could be installed on cars are also being developed. They will significantly increase the functionality of quadcopters in solving monitoring tasks for the housing and utilities sector and transport infrastructure, monitoring the environmental situation and ensuring public safety, and in the case of violations of land and urban planning legislation.&nbsp;</p><hr><p>Translated by <a href="https://news.itmo.ru/en/profile/51/" rel="noopener noreferrer" target="_blank">Catherine Zavodova</a> and originally published by <a data-saferedirecturl="https://www.google.com/url?q=https://news.itmo.ru/en/&source=gmail&ust=1586442733106000&usg=AFQjCNG_vhs0a3Fgz4TwwosC8jzXckCVCQ" href="https://news.itmo.ru/en/" target="_blank">ITMO.NEWS</a></p>

Robots are gradually permeating into more and more areas of our daily life: smart cars are at work on storage facilities and production sites, they test delivery services.

Quadcopter vs. Traffic Safety Inspectors: Researchers Test Drones for Traffic Road Accident Analysis

<p>Researchers at the University of Michigan have developed an ultra-energy-efficient processor to help autonomous micro aerial vehicles navigate their environment. They were able to take advantage of highly effective deep learning algorithms to realize visual simultaneous localization and mapping (SLAM) technology in a single chip measuring a little more than a centimeter.</p><p>The goal of SLAM is to provide a continual estimate of where a moving, autonomous device is located is located along its posture within its environment, both vertically and horizontally, while also constructing a 3D map as the robot navigates its environment. SLAM, which is also used extensively in augmented/virtual reality applications, can use a variety of sensing and location techniques such as lidar, GPS, and cameras.</p><p>Visual SLAM relies solely on cameras, and is the preferred method for autonomous aerial vehicles. However, cameras require extensive processing, which increases the amount of power needed and the overall weight for the battery. That added weight is incompatible with micro aerial vehicles &ndash; some of which can be mere centimeters.</p><p><img src="https://s3-us-west-1.amazonaws.com/wevolver-project-images/froala%2F1575288164203-slam-system-content.jpg" style="width: 1114px;" class="fr-fic fr-dib"></p><p><em>SLAM system.&nbsp;</em></p><p>A team of faculty and students, led by Professors Hun-Seok Kim, David Blaauw, and Dennis Sylvester, have created the first visual SLAM processor on a single chip that provides highly accurate, low-power, and real-time results.</p><p>&ldquo;Our processor can be integrated into the camera so that as the camera moves, the processor continuously computes its location and posture, while accurately mapping the surrounding area,&rdquo; said Ziyun Li, a doctoral student working on the project.</p><p>The researchers were able to do this using convolutional neural networks (CNNs), a deep learning algorithm that has proven to be highly successful in image analysis. However, since CNNs require significant resources for computation and memory, they first needed to develop hardware that could support a CNN while flying onboard a drone.</p><p>Similar systems developed by others had resulted in an incomplete SLAM system, or one that was limited in quality and range. A more recent attempt required an off-chip high-precision inertial measurement unit to simplify the complexity that comes with using CNN, which increased both the cost and the power.</p><p>&ldquo;We implemented a single chip system,&rdquo; said Li, &ldquo;that allowed us to use the computation-heavy and accurate CNN algorithm to identify features.&rdquo; The system also accounted for drift <em>(i.e.,</em> small errors that accumulate when mapping a path), resulting in a highly accurate map of the surrounding area.</p><p><span class="fr-video fr-deletable fr-fvc fr-dvb fr-draggable" contenteditable="false" draggable="true"><iframe src="https://www.youtube.com/embed/3xgf5qWfGgk?&wmode=opaque" frameborder="0" allowfullscreen="" class="fr-draggable" style="width: 1115px; height: 624px;"></iframe></span></p><p><em>This video shows the Michigan CNN-SLAM processor in action. The incoming image stream is processed and the trajectory is estimated in real-time. The estimated path (~1km range) can be used for localization and navigation of the car.&nbsp;</em></p><p>The system was tested on the industrial standard KITTI benchmark, which is often used to compare the vision systems used in autonomous vehicles. It achieved 97.9% accuracy in translation and 99.34% in rotation on KITTI rendering automotive scenes over 1km. The 10.92mm<sup>2</sup> chip consumed 243.6mW to process 80fps VGA at 3.6TOPS/W, marking a 15x improvement in performance and 1.44x in energy efficiency over existing systems.</p><p>Future applications of the technology include facilitating human computer interactions, such as virtual reality scenarios where changing the position of your head signals a specific intention.</p>

The SLAM system onboard a drone in the atrium of the EECS Building.

Researchers built the first visual SLAM processor on a single chip that provides highly accurate, low-power, and real-time results.

SLAM-ming good hardware for drone navigation

<p class="summary drop-cap">Anyone who&rsquo;s seen a flock of starlings twist and turn across the sky may have wondered: How do they maneuver in such close formation without colliding?</p><p>&ldquo;Many types of animals swarm or flock or otherwise move in coordinated ways,&rdquo; says <a href="https://profiles.stanford.edu/nicholas-ouellette" target="_blank">Nicholas Ouellette</a>, associate professor of civil and environmental engineering at Stanford. &ldquo;No individual animal knows what every other animal is doing, yet somehow they move cohesively as a group.&rdquo;</p><p>Understanding precisely how they do this, Ouellette says, may be a key to helping engineers design &ldquo;flocks&rdquo; of aerial drones and driverless cars. Emulating animal swarms is attractive because they not only operate without central control but they&rsquo;re also fault-tolerant, to use an engineering term, meaning they can adapt quickly and gracefully to sudden or unexpected conditions. Swarms are also resilient in that they can operate in dirty, disturbed environments.</p><p>So Ouellette and his team have embarked on a series of studies that explore how animal swarms develop the kind of self-organized and self-regulated systems that would allow devices such as drones and mobile sensors to operate safely and efficiently, without the sort of top-down controls that typify, for example, something like the air traffic control system.</p><p>Rather than examine starlings or other birds, which would be impractical to study in the lab because of the space they need to fly, Ouellette and his crew look at non-biting midges, a type of flying insect often found near water or in the shade of trees. Unlike starlings, whose graceful formations, called murmurations, make us stare up at the sky, midges buzz around in chaotic, cloud-like masses. But, says Ouellette, midge swarms still share common features with bird flocks and other animal groups, in that the swarms stay cohesive without any outside control or leadership.</p><p>&ldquo;Midges are so small, they&rsquo;re easy to keep in a lab,&rdquo; Ouellette says. &ldquo;They swarm at dusk and dawn and are easy to cue with lights.&rdquo; In his lab, the midges live in a roughly 5-foot plastic cube, surrounded by high-speed cameras that he uses to reconstruct 3D trajectories for each insect &mdash; direction, acceleration &mdash; all the essential kinematic information needed to characterize swarm movements.</p><p>Working in the lab allows Ouellette&rsquo;s team to run experiments to test different models. The team starts with observed data taken from the camera images. Then they use this data to test hypotheses about what sorts of rules might govern midge movements. But because natural swarm behavior does not provide enough information to distinguish between many hypotheses, the researchers also disrupt the swarm using light and sound to observe whether the real swarm is affected as their proposed rules had predicted.</p><p>Ouellette says it&rsquo;s too early to design artificial systems that behave the same way as animal groups. But his team is developing the sort of experimental environment that other researchers may one day use to build and test high-tech systems based on the bottom-up wisdom of the swarms rather than the top-down rules that typify many technological undertakings.</p><p>&ldquo;As engineers we like to control things,&rdquo; Ouellette says. &ldquo;But we have a lot to learn from insects and other animals that operate just fine without human command and control frameworks.&rdquo;</p>

A starling roost in the United Kingdom. | Unsplash/James Wainscoat

Civil engineers observe how flocks of animals avoid collisions to help devise safe, self-regulated systems for drones and driverless cars.

Could studying how animals move cohesively in a group teach us how to help drones fly safely?

<p>Drones are being used for a growing number of purposes. Their designs are ever more efficient, and techniques for flying them are being further refined all the time. But drones still have the same weak point: their battery. This is particularly true of propeller drones, which are popular for information-gathering purposes in dangerous or hard-to-reach regions. These drones can fly for only around 15 minutes at a time because their engines quickly burn through their batteries. One way of addressing this limitation &ndash; without weighing the drones down &ndash; would be to recharge them while aloft using a power beaming system: an energy-rich laser beam that is guided by a tracking system and shines directly on photovoltaic cells on the drones&rsquo; exterior.</p><p>Several labs around the world, including in the US, have been working on this idea in recent years. LakeDiamond, an EPFL spin-off based at Innovation Park, has now demonstrated the feasibility of using a high-power laser for this purpose. What&rsquo;s more, LakeDiamond&rsquo;s laser emits a wavelength that cannot damage human skin or eyes &ndash; the issue of safety is paramount, since the system is meant for use with civilian drones. LakeDiamond&rsquo;s technology is built around diamonds that are grown in the company&rsquo;s lab and subsequently etched at the atomic level.</p><p><strong>World record for power</strong></p><p>Despite appearances, standard laser beams are not as straight as they seem: as they travel, they expand ever so slightly, leading to a loss in density as they go. But LakeDiamond&rsquo;s system produces a laser beam with a wavelength of 1.5 &micro;m that, in addition to being safe, can travel much farther without losing strength. &ldquo;Systems developed by other companies and labs, often for military applications, employ lasers that are more powerful and thus more dangerous for humans,&quot; says Pascal Gallo, CEO of LakeDiamond. His company took the opposite tack: their technology transforms the rays emitted by a simple low-power diode into a high-quality laser beam. Their beam has a larger diameter, and its rays remain parallel over a longer distance &ndash; in this case up to several hundred meters.</p><p>In LakeDiamond&rsquo;s laser, the light produced by a diode is directed at a booster composed of reflective material, an optical component and a small metal plate to absorb the heat. The breakthrough lies not with this set-up, which already exists, but with the fact that the emitted beam is only a few dozen watts strong. The secret is using a small square lab-grown diamond as the optical component, as this delivers unparalleled performance. LakeDiamond&#39;s system holds the world record for continuous operation using a wavelength in the middle of the infrared range &ndash; it delivers more than 30 watts in its base configuration. &ldquo;That&rsquo;s equivalent to around 10,000 laser pointers,&rdquo; adds Gallo.</p><p><img src="https://s3-us-west-1.amazonaws.com/wevolver-project-images/froala%2F1571294491761-1920x1080+%282%29.jpg" style="width: 766px;" class="fr-fic fr-dib"></p><p>The lab-grown diamonds&rsquo; key properties include high transparency and thermal conductivity. Achieving those things &ndash; and mastering the nano-etching process &ndash; took the researchers over ten years of development. LakeDiamond grows its diamonds through a process of chemical vapor deposition, an approach that ensures their purity and reproducibility. The surfaces of the resulting tiny square diamonds are then sculpted at the nano level using expertise developed in Niels Quack&#39;s lab at EPFL (<u><a href="https://actu.epfl.ch/news/carving-diamonds-for-optical-components/" target="_blank">read the EPFL article on this topic</a></u>). Thanks to their inherent properties and etched shapes, the diamonds are able to transfer heat to a small metal plate that dissipates it, while at the same time reflecting light in such a way as to create a laser beam.</p><p>&ldquo;To achieve greater power &ndash; say to recharge a larger drone &ndash; these lasers could easily be operated in series,&rdquo; says Nicolas Malpiece, who is in charge of power beaming at LakeDiamond. The company&rsquo;s remote recharging system works in the lab but will require further development and refinement before it&rsquo;s ready for field use. What would happen if a drone flies behind an obstacle and is cut off from its laser energy source? Several approaches to this problem are currently being explored. A small back-up battery could take over temporarily, or, for information-gathering missions over rough terrain for example, the drone could simply return to within range of the laser in order to top up its battery.</p><p>This energy transmission system is also interesting for other areas of application. It can for example be used for charging and transmitting data to satellites. The development of the system is included in a support program of the Swiss Space Office, which began on 1 November and runs for two years.</p>

A small lab-grown diamond measuring a few millimeters per side could one day enable civilian drones to be recharged in mid-flight through a laser. Thanks to the diamond, the laser beam can remain strong enough over a long distance to recharge photovoltaic cells on the drones’ surface. This system, which poses no threat to human health, is being developed by EPFL spin-off LakeDiamond. It could also be used to transmit both power and data to satellites and has just been included in the ten projects supported for two years by of the Swiss Space Office.

Drones are being used for a growing number of purposes. Their designs are ever more efficient, and techniques for flying them are being further refined all the time.

Using diamonds to recharge civilian drones in flight

<p>To build more aerodynamic machines, researchers are drawing inspiration from an unlikely source: the ocean.</p><p>A team of evolutionary biologists and engineers at Harvard University, in collaboration with colleagues from the University of South Carolina, have shed light on a decades-old mystery about sharkskin and, in the process, demonstrated a new, bioinspired structure that could improve the aerodynamic performance of planes, wind turbines, drones and cars.</p><p>The research is published in the <a href="http://rsif.royalsocietypublishing.org/content/15/139/20170828" target="_blank">Journal of the Royal Society Interface</a>.</p><p>Sharks and airplanes aren&rsquo;t actually all that different. Both are designed to efficiently move through fluid (water and air), using the shape of their bodies to generate lift and decrease drag. The difference is, sharks have about a 400 million-year head start on the design process.</p><p><img src="https://s3-us-west-1.amazonaws.com/wevolver-project-images/froala%2F1569467990398-Figure_1+Sharkskin.jpg" style="width: 766px;" class="fr-fic fr-dib"></p><p class="inline_caption"><em>Environmental scanning electron microscope image of denticles from the shortfin mako shark (a) and (its corresponding parametric 3D model (b). These denticles were arranged in a wide range of different configurations on an aerofoil, two examples of which are shown here (c,d ). (Image courtesy of Harvard University)</em></p><p>&ldquo;The skin of sharks is covered by thousands and thousands of small scales, or denticles, which vary in shape and size around the body,&rdquo; said <a href="http://www.people.fas.harvard.edu/~glauder/lauder.htm" target="_blank">George Lauder</a>, the Henry Bryant Bigelow Professor of Ichthyology and Professor of Biology in the Department of Organismic and Evolutionary Biology, and co-author of the research. &nbsp;&ldquo;We know a lot about the structure of these denticles &mdash; which are very similar to human teeth &mdash; but the function has been debated.&rdquo;</p><p>Most research has focused on the drag reducing properties of denticles but Lauder and his team wondered if there was more to the story.</p><p>&ldquo;We asked, what if instead of mainly reducing drag, these particular shapes were actually better suited for increasing lift,&rdquo; said Mehdi Saadat, a postdoctoral fellow at Harvard and co-first author of the study. Saadat holds a joint appointment in Mechanical Engineering at the University of South Carolina.</p><p>To help test that hypothesis, the researchers collaborated with a team of engineers from the <a href="http://seas.harvard.edu/" target="_blank">Harvard John A. Paulson School of Engineering and Applied Sciences</a> (SEAS). &nbsp;For inspiration, they turned to the shortfin mako, the fastest shark in the world. The mako&rsquo;s denticles have three raised ridges, like a trident. Using micro-CT scanning, the team imaged and modeled the denticles in three dimensions. Next, they 3-D printed the shapes on the surface of a wing with a curved aerodynamic cross-section, known as an airfoil.</p><p class="pullquote">The skin of sharks is covered by thousands and thousands of small scales, or denticles, which vary in shape and size around the body.&nbsp;We know a lot about the structure of these denticles &mdash; which are very similar to human teeth &mdash; but the function has been debated.<a class="twitter-quote" href="https://twitter.com/intent/tweet?text=The%20skin%20of%20sharks%20is%20covered%20by%20thousands%20and%20thousands%20of%20small%20scales,%20or%20denticles,%20which%20vary%20in%20shape%20and%20size%20around%20the%20body.%C2%A0We%20know%20a%20lot%20about%20the%20structure%20of%20these%20denticles%20%E2%80%94%20which%20are%20very%20similar%20to%20human%20teeth%20%E2%80%94%20but%20the%20function%20has%20been%20debated." target="_blank">Tweet This</a></p><p>&ldquo;Airfoils are a primary component of all aerial devices,&rdquo; said August Domel, a Ph.D. student at Harvard and co-first author of the paper. &ldquo;We wanted to test these structures on airfoils as a way of measuring their effect on lift and drag for applications in the design of various aerial devices such as drones, airplanes, and wind turbines.&rdquo;</p><p>The researchers tested 20 different configurations of denticle sizes, rows and row positions on airfoils inside a water flow tank. They found that in addition to reducing drag, the denticle-shaped structures significantly increased lift, acting as high-powered, low-profile vortex generators.</p><p>Even if you don&rsquo;t know what a vortex generator is, you&rsquo;ve seen one in action. Cars and planes are equipped with these small, passive devices designed to alter the air flow over the surface of a moving object to make it more aerodynamic. Most vortex generators in the field today have a simple, blade-like design.</p><p>&ldquo;These shark-inspired vortex generators achieve lift-to-drag ratio improvements of up to 323 percent compared to an airfoil without vortex generators,&rdquo; said Domel. &ldquo;With these proof of concept designs, we&rsquo;ve demonstrated that these bioinspired vortex generators have the potential to outperform traditional designs.&rdquo;</p><p>&ldquo;You can imagine these vortex generators being used on wind turbines or drones to increase the efficiency of the blades,&rdquo; said <a href="https://www.seas.harvard.edu/directory/bertoldi" target="_blank">Katia Bertoldi</a>, William and Ami Kuan Danoff Professor of Applied Mechanics at SEAS and co-author of the study. &ldquo;The results open new avenues for improved, bioinspired aerodynamic designs.&rdquo;</p><p>&ldquo;This research not only outlines a novel shape for vortex generators but also provides insight into the role of complex and potentially multifunctional shark denticles,&rdquo; said Lauder.</p>

Environmental scanning electron microscope (ESEM) image of denticles from the shortfin mako shark (scale bar: 200 mm) (Image courtesy of Harvard University)

Bioinspired vortex generators increase airfoil lift, decrease drag

Using shark scales to design better drones, planes, and wind turbines

<div itemprop="description"><p>The main drawback of small drones is their limited flight time, which results from their heavy batteries and relatively inefficient propellers. So for longer missions, a small drone’s battery has to be regularly recharged or replaced. While engineers are working on new designs to address this problem, no real solution has been found – until now. Samir Bouabdallah, who has over 20 years of experience developing drone technology at EPFL and ETH Zurich, has come up with an inventive propulsion system modeled after those used by helicopters. His design, marketed through his startup Flybotix, uses just two propellers and an algorithm-based stabilization mechanism, giving his drones “the aerodynamic performance of a helicopter and the mechanical stability of a quadcopter.” Recently in China for the CES Asia conference and the Venture Leaders China program, Bouabdallah has decided to explore opportunities in this promising market.</p><p><span class="fr-video fr-fvc fr-dvb fr-draggable" contenteditable="false" draggable="true"><iframe src="https://www.youtube.com/embed/kxfCGZ-pN3E?&amp;wmode=opaque" allowfullscreen="" class="fr-draggable" style="width: 766px; height: 430px;" frameborder="0"></iframe></span></p><p><strong>Two rotors guided by stabilization algorithms&nbsp;</strong></p><p>A drone’s flying time is proportional to its size – the smaller the machine, the less time it can spend in the air. “Drones with four rotors are highly stable, making them reliable and easy to use,” says Bouabdallah. But as these drones get smaller, their propellers become increasingly inefficient. That, combined with their limited battery capacity, explains their abbreviated flying times.</p><p>One way to extend those times is to use a completely different kind of propulsion system. By reducing the number of propellers from four to two, for instance, engineers can make each propeller longer and the rotor more efficient. That also cuts the amount of power required, meaning the same-sized battery can run for longer. But until now, drones with two propellers were generally less stable and harder to manipulate. Bouabdallah’s design employs a ring-like structure – around 30 centimeters in diameter – in which the propellers are stacked on top of each other in the center and turn in opposite directions. His drones look more like flying saucers than the machines we’re used to seeing.</p><p><img src="https://s3-us-west-1.amazonaws.com/wevolver-project-images/froala%2F1564073300650-Flybotix_drone.jpg" style="width: 766px;" class="fr-fic fr-dib"></p><p>Another innovation relates to his drones’ stabilization mechanism. Helicopters use a complicated transmission system that enables pilots to control the blades and maintain the aircraft’s pitch. But this system is heavy, cumbersome and requires a considerable amount of maintenance. However, Bouabdallah has developed an algorithm that offsets the two rotating forces and serves the same function as a helicopter’s transmission system. His drones can therefore be piloted just as easily as quadcopters, using a conventional remote-control device.</p><p>Bouabdallah’s drone, with its unique transmission system and foam covering its outer ring, can bounce off of obstacles without breaking. The first application that Flybotix will target is inspecting dangerous or hard-to-reach areas. While in China last month the team got a lot of positive feedback on their technology and made some promising contacts. They hope to soon launch their drones on the Chinese market.</p></div>

Flybotix has developed a novel drone with just two propellers and an advanced stabilization system that allow it to fly for twice as long as conventional models. That fact, together with its small size, makes it perfect for inspecting hard-to-reach parts of industrial facilities such as ducts.

unmanned aerial vehicles (UAVs)

ORGANIZATIONS.

SHAPING THE INDUSTRY.

InDro Robotics

SAE International

Cloud Ground Control

FLIR

Latest Posts