All posts by Rens van Poppel

Yes, you heard that correctly: the goal is permanent airtime. Robotic flies roaming a room in RoboHouse with no human guidance – achieved within six months. In the future, 24/7 swarms like these may revolutionise aircraft inspection. Imagine a fighter jet enveloped by hundreds of nano drones that build-up a detailed picture in minutes. It’s a challenging mission, but not all challenges are equal. So we asked each Crazyflies team member: What is your favourite problem?

Lennart #myfavouritedesignproblem

Okay, maybe permanent flying is exaggerating a bit, at some point batteries need recharging, but it remains the overall design essence. For team member Lennart, this is the main challenge: “We want to optimise the charging process so that you have as many drones in the air as possible with a minimum amount of charging pads.”

Each Crazyflie can buzz off for seven minutes before needing a 35 minute recharge. Through the use of wireless charging pads, human intervention is cancelled out, the alternative being manual battery replacement.

Seppe #myfavouritedesignproblem

But challenges go way further than just battery strategy. Student Seppe identifies his favourite obstacle-to-overcome in collision avoidence: “This does not only include collisions between drones, but also with stationary objects,” Seppe tells us. “By deploying sensors and proper coding, these risks are minimised. Yet the strength of a robust system doesn’t lie in reducing risks, it lies in handling them when they happen.”

Servaas #myfavouritedesignproblem

Servaas’s favourite challenge ties in with that of his colleague: round-trip latency. Or in English: the time it takes for the flying AI-insects to send their observations and receive commands in return. “Depending on how much time this transfer of information takes up, we could for instance let the drones react to more unpredictable objects such as humans.” Perhaps actual flies could also identify as such an object.

Andreas #myfavouritedesignproblem

Floating away from technical aspects, Andreas defines solving real-world problems his goal: “Designing an autonomous, 24/7 flying drone swarm is cool, but we also want to have an actual impact through real-world application.” Andreas seeks to fulfil this wish by doing market research and identifying problems that yet remain devoid of a solution. One such application could be the inspection of large or difficult-to-access infrastructure like bridges or power lines.

Andrea #myfavouritedesignproblem

Not coming from a robotic background, for fifth team member Andrea the challenge amounted to familiarising all this software involved. Luckily, Andrea managed to learn the tools of the trade, finding the AI-insects’ autonomy one of the next exciting challenges to be tackled.

The drones

But wait, this does not yet complete the team. There are a hundred other individuals, quite literally also team members. The students have included the Crazyflies in their team, deciding to name them ‘member 6 to 105’. These drones are going to inspect infrastructure all by themselves, only stopping occasionally to recharge their batteries.

Cyberzoo

If all goes well, the Crazyflies could become part of the Crazy Zoo robot exhibition on TU Delft Campus, an initiative by Chris Verhoeven, theme leader swarm robots at TU Delft. For now though, the students have a lot of work on their hands to realise their dreams and live up to the challenges. We have no doubt they will fly high.

The post Robotic flies to swarm 24/7 in RoboHouse appeared first on RoboHouse.

The final episode of our RoboHouse Interview Trilogy: ‘The Working Life of the Robotics Engineer’ interviews Srimannarayana Baratam. Sriman, as he is also called, co-founded the company Perciv.ai just two months after graduating. Rens van Poppel explores his journey so far.

Perciv.ai claims that AI-driven machine perception could become affordable to everyone. When was this vision formed, and how did it come about? Sriman points to the period right after his graduation. He says it was pivotal for building trust with partners, and consensus with effective communication. Because starting your own company comes with a lot of challenges.

 

“It is important for to find partners you can trust,” says Sriman. “You need to understand each other’s motivation and commitment. You need to assess what real value does this person add to the team.”

Coming from an automotive background in India, Sriman’s master’s thesis investigated the use of radar and cameras to protect vulnerable people in urban environments. He co-founded the start-up with his supervisor, Dr András Pálffy, and Balazs Szekeres, another robotics student who heard about the project. In the two months after his graduation, Sriman and his co-founders came together to focus full-time on their vision for Perciv.ai.

“In July and August we sat down and discussed the vision between the three of us,” he says. This period also led to tough conversations, ranging from finance to market strategy. “When finally the main questions were sorted out, you just got to take that leap of faith together. This leap of faith proved fruitful, seeing that the high level of trust resulted in a high level of productivity over the past five months.”

Since then Perciv.ai went on to win the NWO take-off phase 1 grant, got their own office and workspace in RoboHouse, signed a contract with an unmanned aerial vehicle (UAV) company and in doing so, generated their first sales revenue.

“This does not mean that there are no more heavy debates,” Sriman adds. “We all share the same vision, but in order to reach our goal of a sustainable and affordable product, we sometimes have different ideas on what that final product should look like.”

Sriman’s passion for robotics and the company’s goals is palpable: “We want to make machine perception technology available to all.”

The post RoboHouse Interview Trilogy, Part III: Srimannarayana Baratam and Perciv.ai appeared first on RoboHouse.

For the second part of our RoboHouse Interview Trilogy: The Working Life of the Robotics Engineer we speak with Wendel Postma, chief engineer at Project MARCH VIII. How does he resolve the conundrum of integration: getting a bunch of single-minded engineers to ultimately serve the needs of one single exoskeleton user? Rens van Poppel inquires.

Wendel oversees technical engineering quality, and shares responsible for on-time delivery within budget with the other project managers. He spends his days wandering around the Dream Hall on TU Delft Campus, encouraging his team to explore new avenues for developing the exoskeleton. What is possible within the time that we have? Can conflicting design solutions work together?

Bringing bad news is part of the chief engineer’s job.

Project MARCH is iterative enterprise.

Most of its workplace drama comes from the urgency to deliver at least one significant improvement on the existing prototype. This year’s obsessions is weight; a lighter exoskeleton would require less power from both pilot and motors. Self-balancing would become easier to realise.

In order not to weaken the frame of the exoskeleton, there was a lot of enthusiasm to experiment with carbon fibre, which is both a light and strong material. Something, however, got in the way: the team struggled to find a pilot.

My job is making sure that in the end we don’t have 600 separate parts, but one exoskeleton.

“Having a test pilot is crucial if we are to reach our goals,” Wendel says. “Our current exoskeleton is built to fit the particular body shape of the person controlling it. The design is not yet adjustable to a different body shape. So it is crucial to get the pilot involved as quickly as possible.”

Not having a pilot was stressful for the entire team.

Their dream of creating a self-balancing exoskeleton was in danger. Wendel had to step up: “As chief engineer you have to make tough decisions. Carbon fibre is strong, but not flexible and difficult to machine. That is why we switched to aluminium, because it is easier to modify even after it is finished.”

“It was a huge disappointment,” Wendel says. “Some of us had already finished trainings for carbon manufacturing. Carbon parts were already ordered. The team felt let down. We had spent a so much time on something that was now impossible – because of the delays caused by having no pilot.”

“I learnt that bringing bad news is part of the chief engineer’s job. The next step is to look at how to convert the engineers’ enthusiasm for carbon fibre into new solutions and to redeploy their personal qualities.”

Wendel says the job also taught him to consider a hundred things at the same time. And to make sacrifices. Project MARCH involves long workdays and maybe not seeing your friends and roommates as much as you would like.

As a naturally curious person, Wendel found out that curiosity must be complemented by grit to make it in robotics. You often need to go deeper and study in more detail to make a good decision. “It is hard work. However, that is also what makes the job so much fun. You work in such a highly motivated team.”

That is also what makes the job so much fun.

The carbon story ended well, though.

When the team did found a pilot, hard-working Koen van Zeeland, the choice for aluminium as a base material paid off. Through a process of weight analysis, parts can now be optimised for an ever lighter exoskeleton.

The Project MARCH team continues to grow through setbacks and has doubled-down on their efforts to create the world’s first self-balancing exoskeleton. If they succeed, it will be a huge success for this unique way of running a business.

The post RoboHouse Interview Trilogy, Part II: Wendel Postma and Project MARCH appeared first on RoboHouse.

Part one of our RoboHouse Interview Trilogy: The Working Life of Robotics Engineers seeks out Christian Geckeler. Christian is a PhD student at the Environmental Robotics Lab of ETH Zürich. He speaks with Rens van Poppel about the experience of getting high into the wild.

What if drones could help place sensors in forests more easily? What if a sensor device could automatically grab and hold a tree branch? Which flexible material is also strong and biodegradable? These leaps of imagination lead Christian to a new kind of gripper, inspired by the Japanese art of folding.

His origami design wraps itself around tree branches close enough to trigger an unfolding movement. This invention may in the future improve our insight into hard-to-access forest canopies, in a way that is environmentally friendly and pleasant for human operators.

What is it like to work in the forest as a researcher with this technology?
“Robotic solutions deployed in forests are currently scarce,” says Christian. “So developing solutions for such an environment is challenging, but also rewarding. Personally I also enjoy being outdoors. Compared to a lab, the forest is wilder and more unpredictable. Which I find wonderful, except when it’s cold.”

Are there limits as to where the gripper can be deployed?
“The gripper is quite versatile. Rather than the type of trees, it is the diameter and angle of the branch that dictate whether the gripper can attach. Even so, dense foliage could hinder the drone, and there should be sufficient space for the gripper to attach.”

Are the used materials environmentally friendly?
“Currently not all components are biodegradable, and the gripper must be recollected after sampling is finished. However, we are currently working on a fully biodegradable gripper, which releases itself and falls on the ground after being exposed to sufficient amounts of water, which makes collection much easier.”

How good at outdoor living do aspiring tree-canopy researchers need to be?
“Everything is a learning process,” says Christian philosophically. “Rather than existing expertise, a willingness to learn and passion for the subject is much more important.”

What happens when the drone gets stuck in a tree?
“As a safety measure, the drone has a protective net on top which prevents leaves and branches from coming in contact with the propeller. And we avoid interaction between the drone and foliage, so this has never happened.”

What struck you when took the gripper into the wild?
“Perhaps the most surprising thing was the great variance that is found in nature; no two trees are alike and every branch is different. The only way of finding out if your solution works is by testing outside as soon and as often as possible.”

Christian ends with a note on the importance of social and technical interplay in robotics: “You may think you develop a robot perfectly, but you must make sure society actually wants it and that it is easy to use for not technically-minded people too.”

The post RoboHouse Interview Trilogy, Part I: Christian Geckeler and The Origami Gripper appeared first on RoboHouse.