ETH Zurich researchers led by robotics professor Marco Hutter have developed a new control approach that enables a legged robot, called ANYmal, to move quickly and steadily over difficult landform. Thanks to machine learning, the robot can combine its visual perception (感知) of the environment with its sense of touch for the first time.
Steep sections on slippery ground, high steps, stone and forest trails full of roots: the path up the 1098-metre-high Mount Etzel at the southern end of Lake Zurich is filled with masses of barriers. But ANYmal, the legged robot from the Robotic Systems Lab at ETH Zurich, overcomes the 120 vertical metres effortlessly in a 31-minute hike. That’s 4 minutes faster than the evaluated duration for human hikers and with no falls or missteps.
This is made possible by a new control technology, which researchers at ETH Zurich led by Marco Hutter recently presented in the journal Science Robotics. “The robot has learned to combine visual perception of its environment with its sense of touch based on direct leg contact (接触). This allows it to cope with rough landform faster, more efficiently and, above all, more steadily,” Hutter says.
Before the robot could put its abilities to the test in the real world, Marco Hutter exposed the system to masses of barriers and sources of error in a virtual training camp. This let the network learn the perfect way for the robot to overcome barriers, as well as when it can rely on environmental data and when it would do better to ignore that data. “With this training, the robot is able to master the most difficult natural landform without having seen it before,” says ETH Zurich Professor Hutter.
In the future, ANYmal can be used anywhere that is too dangerous for humans or too impassable for other robots. Whether after an earthquake, after a nuclear disaster, or during a forest fire, robots like ANYmal can be used primarily wherever it is too dangerous for humans and where other robots cannot cope with the difficult landform.
【小题1】How was ANYmal’s 120-vertical-metre hike in the end?| A．Tough. | B．Successful. | C．Challenging. | D．Attractive. |
| A．By analyzing reasons. | B．By making comparisons. |
| C．By listing data. | D．By giving examples. |
| A．It had no trouble in a long hike. | B．It has been applied to test landform. |
| C．It needed tests before being put into use. | D．It overcame barriers based on indirect leg contact. |
| A．Disasters’ severity. | B．Humans’ limitation. |
| C．ANYmal’s future appearance. | D．ANYmal’s development potential. |
