Lynx S10: Robot dog travels autonomously on Arctic ice floes for the first time

Chinese robotics manufacturer Deep Robotics has modified its autonomous four-legged robot Lynx S10 so that it can function in extreme weather conditions and move autonomously in unpredictable terrain formations made of meltwater, snow and ice. The company tested the robot on an Arctic expedition, during which it also had to autonomously navigate ice floes.

As standard, the Lynx S10 has four legs to which four wheels are attached. With this combination, the robot can switch between walking and driving and select the most advantageous type of movement depending on the surface. However, this is not enough to cope with extreme terrain conditions such as those in the Arctic. Because the robot could get stuck with its wheels on the changing surfaces between ice, snow and meltwater puddles.

The engineers at Deep Robotics have therefore replaced the wheels with large biomimetic paws based on those of polar bears. Thanks to their width, the paws enable better weight distribution over a larger area and prevent them from sinking into soft snow. The artificial paws also have a non-slip surface and integrated crampons in the form of claws, which enable better traction on hard ice.

The developers increased the surface area of ​​the robot’s limbs, which can act like a kind of paddle. To help him move around in the water, they gave him small, air-filled water wings on his upper legs. With the legs and feet modified in this way, the Lynx S10 can move better even in environments where ice and water mix.

In addition to the locomotion, the engineers also adapted the robot’s already robust design to the extreme Arctic conditions. Although the robot is dust- and water-resistant as standard and can withstand temperatures between -20 and 50 degrees Celsius, this is not enough for arctic conditions. Accordingly, the sealing of the robot was increased from IP66 to IP67 to ensure better protection against ingress of water. Additionally, cold protection was installed.

Autonomous in an arctic environment

In order to be able to move autonomously in unpredictable terrain, the robot is equipped with autonomous path planning and obstacle detection. With AI-supported motion control and a special gear selection algorithm, the Lynx S10 switches to the optimal form of motion depending on the terrain formation. To recognize the environment, the robot uses four HDR-capable ultra-wide-angle cameras, which provide images with a high contrast range and high image sharpness for evaluation. There are also lidar sensors on the front and back to create detailed three-dimensional maps. The combination of sensors allows the robot to detect obstacles, plan routes and thus find its way without human control. In order to be able to orientate better, the four-legged friend can also stand up more to get an overview from an elevated position.

Deep Robotics tested the robot, which weighs around 20 kg, under uncontrolled conditions as part of an Arctic expedition. The robot was able to cope with the terrain of ice, snow and meltwater – although sometimes only very slowly. The robot also conquered surfaces that consisted of ice mixed with water.

Deep Robotics emphasizes that the modified Lynx S10 is not yet a finished product. The robot is still in the alpha stage. The finished version could be used to support researchers or rescue workers in their work in inhospitable areas that are dangerous, remote and inaccessible to humans.

(olb)