Editor choice


A new era for multi-robot research: Cambridge's RoboMaster

In the quiet halls of the University of Cambridge, a robotic revolution is underway. While the world's attention has been captivated by humanoid robots and self-driving cars, a team of researchers has been quietly working on something that could have an even more profound impact: swarms of intelligent, cooperating robots.



Led by Amanda Prorok, Principal Investigator at Cambridge's Department of Computer Science and Technology, the team has introduced the Cambridge RoboMaster—a groundbreaking platform that promises to transform the field of multi-robot research. Their vision? A future where teams of robots work together seamlessly, tackling complex tasks that are beyond the reach of individual machines.

"Our mission is to develop solutions for collective intelligence in multi-robot and multi-agent systems," Prorok explains. Her ambitions extend far beyond the lab. "This research incorporates methods from machine learning, planning, and control, with a wide range of applications, including automated transport, logistics, environmental monitoring, and search and rescue."

The potential is staggering. Imagine swarms of robots rapidly covering vast distances, simultaneously visiting different sites, or monitoring large geographical areas. Such capabilities could revolutionize fields as diverse as disaster response, warehouse automation, and planetary exploration. But turning this vision into reality requires more than just algorithms—it demands robust, versatile hardware for real-world testing.

Enter the Cambridge RoboMaster. At first glance, these small, wheeled robots might not seem revolutionary. Originally designed for high school and university-level competitions by DJI, they're about the size of a toaster. But in the hands of Prorok's team, they've become something much more.

"This platform provided a solid foundation that we could build upon to meet our specific needs," says Jonas Blumenkamp, a key member of the Cambridge team. Over three years, they've transformed these off-the-shelf robots into state-of-the-art research tools, adding more powerful computers, advanced sensors, and sophisticated control software.

The result is a robot that strikes a perfect balance between size and capability. Unlike smaller robots that often lack computing power, or larger ones too expensive and bulky for indoor use, the Cambridge RoboMaster packs a punch in a compact package. With a top speed of 4.5 meters per second—about as fast as a person running—they're agile enough to navigate complex indoor environments.

But speed is just the beginning. "Our platform includes a control stack for full on-board autonomy, peer-to-peer communication, and can run multi-agent reinforcement learning policies directly from our simulation framework without additional training," Blumenkamp explains. In layman's terms, each robot is not just fast, but smart—capable of making decisions, learning from experience, and communicating with its teammates.

Perhaps most importantly, at around $700 each, the Cambridge RoboMasters are remarkably affordable. This isn't just about saving money; it's about democratizing research. High costs often restrict cutting-edge robotics to a handful of elite institutions. By making their platform accessible, the Cambridge team is inviting universities and labs worldwide to join the multi-robot revolution.

The implications are far-reaching. Today, a robotics lab in Mumbai or Montevideo might struggle to afford a single high-end robot. Tomorrow, with the Cambridge RoboMaster, they could be running experiments with entire robot teams, contributing their unique insights to a global research community.

Already, the Cambridge RoboMasters have proven their mettle in a variety of tests. They've navigated both indoor and outdoor environments on smooth terrains, demonstrating power-efficiency and versatility. In one striking demonstration, five robots moved in tight formation, hinting at future applications where robot teams might need to coordinate closely in confined spaces.

"While we do not intend for this platform to be directly used in real-world settings, it serves as an ideal proxy," Blumenkamp says. "It's a research tool that allows for testing algorithms applicable in multi-agent navigation. Such scenarios are relevant for real-world domains including warehouse automation and logistics."

The Cambridge team isn't keeping their work behind closed doors. They've published all the hardware designs, software, and simulation tools on GitHub, inviting researchers worldwide to start experimenting. This open-source approach could accelerate progress exponentially, as ideas cross-pollinate between labs across the globe.

Looking ahead, the team's ambitions show no signs of slowing. "Right now, we're focusing on improving on-board sensing, decentralized communication, and control," Blumenkamp says. But that's just the start. "We're also looking at how this platform can serve as a bridge for deploying our research to drones."

The vision is a future where the principles of multi-robot cooperation, honed in the controlled environment of the lab, are applied to drones surveying disaster zones or autonomous vehicles navigating city streets. The Cambridge RoboMaster, in this light, isn't just a research tool; it's a stepping stone to a world where machines work together as fluidly as a flock of birds or a school of fish.

"Maybe, one day we will see these robots playing soccer," Blumenkamp muses. It's a playful suggestion, but the underlying message is profound. If robots can learn to collaborate in the dynamic, unpredictable environment of a soccer match, there may be no limit to what they can achieve in the real world.

In the grand narrative of technological progress, breakthroughs often come from unexpected places. As the world watches giants like Boston Dynamics and Tesla, a small team in Cambridge is quietly laying the foundations for a cooperative robotic future. Their tool? A fleet of customized, $700 robots, each no bigger than a toaster, yet each a harbinger of a transformative era in robotics.

Share with friends:

Write and read comments can only authorized users