Assemble Your (Virtual) Robotic Underground Exploration Team

It’s amazing how many things have managed to move online in recent weeks, many with a beneficial side effect of eliminating travel making them more accessible to everyone around the world. Though some events had a virtual track before it was cool, among them the DARPA Subterranean Challenge (SubT) robotics competition. Recent additions to their “Hello World” tutorials (with promise of more to come) have continued to lower the barrier of entry for aspiring roboticists.

We all love watching physical robots explore the real world, which is why SubT’s “Systems Track” gets most of the attention. But such participation is necessarily restricted to people who have the resources to build and transport bulky hardware to the competition site, which is just a tiny subset of all the brilliant minds who can contribute. Hence the “Virtual Track” which is accessible to anyone with a computer that meets requirements. (64-bit Ubuntu 18 with NVIDIA GPU) The tutorials help get us up and running on SubT’s virtual testbed which continues to evolve. With every round, the organizers work to bring the virtual and physical worlds closer together. During the recent Urban Circuit, they made high resolution scans of both the competition course as well as participating robots.

There’s a lot of other traffic on various SubT code repositories. Motivated by Bitbucket sunsetting their Mercurial support, SubT is moving from Bitbucket to GitHub and picking up some housecleaning along the way. Together with the newly added tutorials, this is a great time to dive in and see if you want to assemble a team (both of human collaborators and virtual robots) to join in the next round of virtual SubT. But if you prefer to stay an observer of the physical world, enjoy this writeup with many fun details on systems track robots.

A Hackable Drone Without All The Wiring

Drones have come a long way in the past decade, and a lot of the pioneering work that made them mainstream was done by individual hackers and small teams. This often involves cobbling together components into flying crow’s nests of wiring. To streamline things a bit for hackers, the team at Luminous Bees are working on Ardubee, a small 3″ drone designed from the ground up for hacking.

The Ardubee is built around a single PCB that also acts as the frame of the drone. Onboard is an STM32F427 microcontroller, IMU, barometer and compass, ESCs, ESP8266 for telemetry, and a downward-facing range finder. It’s ready to connect to an SBUS RC receiver and a range of pluggable modules are in development to expand the drone’s capabilities. It’s designed to run the open-source Ardupilot software, which we’ve seen in so many DIY autonomous vehicles. Power is provided by a single 18650, which will probably limit higher speed maneuverability a bit but should be fine for the slower precision flight that such a drone is likely to be used for.

The team already has a swarm of larger 5″ drones that they developed for light shows. In the process they developed their own Ultrawide-band indoor positioning system, which will also be available for the Ardubee. They hope to launch a Kickstarter campaign soon and are asking for input from the community, so they can know what features need to be prioritized. We look forward to seeing where this project goes!

Autonomous vehicles are a popular topic around here for air, land, and water, and we have no doubt there will be many more.

Thanks for the tip [Andreas]!

DARPA Challenge Autonomous Robot Teams To Navigate Unfinished Nuclear Power Plant

Robots might be finding their footing above ground, but today’s autonomous robots have a difficult time operating underground. DARPA wanted to give the state of the art a push forward, so they are running a Subterranean (SubT) Challenge which just wrapped up its latest round. A great review of this Urban Circuit competition (and some of the teams participating in it) has been published by IEEE Spectrum. This is the second of three underground problem subdomains presented to the participants, six months apart, preparing them for the final event which will combine all three types.

If you missed the livestream or prefer edited highlight videos, they’re all part of DARPAtv’s Subterranean Challenge playlist. Today it starts with a compilation of Urban Circuit highlights and continues to other videos. Including team profiles, video walkthrough of competition courses, actual competition footage, edited recap videos, and the awards ceremony. Half of the playlist are video from the Tunnels Circuit six months ago, so we can compare to see how teams performed and what they’ve learned along the way. Many more lessons were learned in the just-completed Urban Circuit and teams will spend the next six months improving their robots. By then we’ll have the Caves Circuit competition with teams ready to learn new lessons about operating robots underground.

Continue reading “DARPA Challenge Autonomous Robot Teams To Navigate Unfinished Nuclear Power Plant”

DARPA Subterranean Challenge Urban Circuit Now Livestreaming

Currently underway is the DARPA Subterranean Challenge (SubT) systems competition for urban circuits streamed live on YouTube now through Wednesday, February 26th.

The DARPA Grand Challenge of 2004 kicked research and development of autonomous vehicles into high gear. Many components on today’s self-driving vehicles can be traced back to systems developed for that competition. Hoping to spur further development, DARPA has since held several more challenges focused moving the state of the art in autonomous robotics ahead.

To succeed in this challenge, robots must handle terrain that would confuse today’s self-driving cars. Cluttered environments, uneven surfaces of different materials, even the occasional flooded section are fair game. These robots also lose access to some of the tools previously available, such as GPS. The “systems track” denotes teams building physical robot systems versus a separate “virtual track” for simulation robots. “Urban circuit” is the second of four phases in this competition, environments of this phase are focused on man-made underground structures. (Think subway station.) For more details on this competition as well as description of various phases, see our introductory post or the competition site.

Those who rather not watch robots tentatively exploring unknown territory (and occasionally failing) may choose to wait for summaries published after competition rounds are complete. The first phase (tunnel circuit) from August-October 2019 was summarized by IEEE Spectrum here. Or you can go straight to DARPA for details on the systems track and virtual track with overall results posted on the competition site.

Continue reading “DARPA Subterranean Challenge Urban Circuit Now Livestreaming”

Making Autonomous Racing Drones Lean And Mean

Recently the MAVLab (Micro Air Vehicle Laboratory) at the Technical University of Delft in the Netherlands proudly proclaimed having made an autonomic drone that’s a mere 72 grams in weight. The best part? It’s designed to take part in drone races. What this means is that using a single camera and onboard processing, this little drone with a diameter of 10 centimeters has to navigate the course, while avoiding obstacles.

To achieve this goal, they took an Eachine trashcan drone, replacing its camera with an open source JeVois smart machine vision camera and the autopilot software with the Paparazzi open UAV software. Naturally, scaling a racing drone down to this size came at an obvious cost: with its low-quality sensors, relatively low-quality camera and limited processing power compared to its big brothers it has to rely strongly on algorithms that compensate for drift and other glitches while racing.

Currently the drone is mainly being tested at a four-gate race track at TU Delft’s Cyberzoo, where it can fly multiple laps at a leisurely two meters per second, using its gate-detecting algorithms to zip from gate to gate. By using machine vision to do the gate detection, the drone can deal with gates being displaced from their position indicated on the course map.

While competitive with other, much larger autonomous racing drones, the system is still far removed from the performance of human-controlled racing drones. To close this gap, MAVLab’s [Christophe De Wagter] mentions that they’re looking at improving the algorithms to make them better at predictive control and state estimation, as well as the machine vision side. Ideally these little drones should be able to be far more nimble and quick than they are today.

See a video of the drone in action after the link.

Continue reading “Making Autonomous Racing Drones Lean And Mean”

DARPA Goes Underground For Next Challenge

We all love reading about creative problem-solving work done by competitors in past DARPA robotic challenges. Some of us even have ambition to join the fray and compete first-hand instead of just reading about them after the fact. If this describes you, step on up to the DARPA Subterranean Challenge.

Following up on past challenges to build autonomous vehicles and humanoid robots, DARPA now wants to focus collective brainpower solving problems encountered by robots working underground. There will be two competition tracks: the Systems Track is what we’ve come to expect, where teams build both the hardware and software of robots tackling the competition course. But there will also be a Virtual Track, opening up the challenge to those without resources to build big expensive physical robots. Competitors on the virtual track will run their competition course in the Gazebo robot simulation environment. This is similar to the NASA Space Robotics Challenge, where algorithms competed to run a virtual robot through tasks in a simulated Mars base. The virtual environment makes the competition accessible for people without machine shops or big budgets. The winner of NASA SRC was, in fact, a one-person team.

Back on the topic of the upcoming DARPA challenge: each track will involve three sub-domains. Each of these have civilian applications in exploration, infrastructure maintenance, and disaster relief as well as the obvious military applications.

  • Man-made tunnel systems
  • Urban underground
  • Natural cave networks

There will be a preliminary circuit competition for each, spaced roughly six months apart, to help teams get warmed up one environment at a time. But for the final event in Fall of 2021, the challenge course will integrate all three types.

More details will be released on Competitor’s Day, taking place September 27th 2018. Registration for the event just opened on August 15th. Best of luck to all the teams! And just like we did for past challenges, we will excitedly follow progress. (And have a good-natured laugh at fails.)

Delivery Drones Can Learn From Driving And Cycling

Increasingly these days drones are being used for urban surveillance, delivery, and examining architectural structures. To do this autonomously often involves using “map-localize-plan” techniques wherein first, the location is determined on a map using GPS, and then based on that, control commands are produced.

A neural network that does steering and collision prediction can compliment the map-localize-plan techniques. However, the neural network needs to be trained using video taken from actual flying drones. But generating that training video involves many hours of flying drones at street level putting vehicles and pedestrians at risk. To train their DroNet, Researchers from the University of Zurich and the Universidad Politecnica de Madrid have come up with safer sources for that video, video recorded from driving cars and bicycles.

DroNet
DroNet

For the drone steering predictions, they used over 70,000 images and corresponding steering angles from the publically available car driving data from Udacity’s Open Source Self-Driving project. For the collision predictions, they mounted a GoPro camera to the handlebars of a bicycle and drove around a city. Video recording began when the bicycle was distant from an object and stopped when very close to the object. In total, they collected 32,000 images.

To use the trained network, images from the drone’s forward-facing camera were fed into the network and the output was a steering angle and a probability of collision, which was turned into a velocity. The drone remained at a constant height above ground, though it did work well from 1.5 meters to 5 meters up. It successfully navigated road lanes and avoided moving pedestrians and bicycles. Intersections did confuse it though, likely due to the open spaces messing with the collision predictions. But we think that shouldn’t be a problem when paired with map-localize-plan techniques as a direction to move through the intersection would be chosen for it using the location on the map.

As you can see in the video below, it not only does a decent job of flying down lanes but it also flies well in a parking garage and a hallway, even though it wasn’t trained for either of these.

Continue reading “Delivery Drones Can Learn From Driving And Cycling”