On a summer day in 2019, Raj Patel and his friend Sally Blake walked past a pool. It was filled with smelly algae (水藻), making it hard for workers to clean up. So the two 17-year-old students from Chicago decided to create a robot to deal with this problem. On Aug 19, the 18^th Science and Environment Invention Award was announced. Patel and Blake’s invention of a water surface cleaning robot received the Gold Award.

It took them about four years to finish the robot. The robot moves like a boat and can be controlled from far away on the waterside. To collect garbage, the robot has two plates as “hands”. Then a conveyor belt is used for sending the garbage to a basket. “There is also a heating sheet under the belt to help dry the garbage and make it easy to deal with,” said Patel.

Compared to other robots of its kind, their robot is small, light and eco-friendly. “It has a solar panel (太阳能电池板) set on top,” said Blake.

The journey to build their robot was full of difficulties. But they worked hard to deal with these problems. One challenge was the balance of the boat. “It used to be heavy and easy to turn over. On the advice of our teacher, we changed the top part from steel to alloy (合金) to make it lighter,” said Patel.

As for their next step, they hope to make the robot work on its own. “We will continue to learn programming skills to make our dreams come true,” said Blake.

Making eye contact with a robot can be a very strange experience. Scientists even have a name for the teeing: the “uncanny valley”. Now, researchers at the Istituto Italiano di Tecnologia (IIT) in Italy have found that it is more than just a feeling. They ran an experiment that showed how a robot’s gaze can trick people into thinking they are socially interacting with a human being. That experience can slow a person’s ability to make decisions.

“Gaze is an extremely important social signal,” said the lead writer, Professor Agnieszka Wykowska, “The question is whether the robot’s gaze will evoke very similar mechanisms in the human brain as a human’s gaze would.”

The team asked 40 participants to play a video game of “chicken”. Each player had to decide whether to permit a car to run straight toward another car or to turn to avoid a crash. The people played against a human-like robot sitting across from them. During breaks in the game, players had to look at the robot. Sometimes the robot would look back and at other times it would look away. As the interactions happened, the scientists collected data on participants’ behavior and brain activity.

“Our results show the gaze of the robot had an impact on the way humans made decisions and humans’ responses so humans were much slower in making decisions in the game,” Wykowska said. Given that the robot, is designed loosely to copy the shape and appearance of people, it’s not altogether startling perhaps that its gaze can influence people’s attention.

The findings could be useful for deciding where and how human-like robots might be placed. When we understand when and how robots change people’s social behavior, “then we can decide in which sort of context this is desirable and beneficial for humans and in which context this should not occur,” Wykowska said.

A tracked robot approaches a pile of brushwood blocking its path. This is Rolan, short for Robot Manipulator (操控者), who is practising for what is its graduation ceremony, when it will show off its skills. After a pause, it reaches out an arm, takes hold of a branch, lifts it up and drags it clear.

Though this is a small action for a human being, it is a breakthrough for robots, according to Stuart Young of the Army Research Laboratory, who is in charge of the RoMan project. As is known, robots easily become confused by something unexpected, some unfamiliar and none with convenient handles. Taking it apart is far beyond the power of any industrial robot. Dr. Young says that, as far as he knows, RoMan is the first machine able to manipulate unfamiliar objects in an unknown environment.

Just as a human being would, it has to learn about the world through observation and experiment before manipulating it. So it is trained on numerous tree branches until it is able to recognize unfamiliar ones for what they are and knows to grasp the trunk, rather than the leaves or the branches. Having grasped an object, RoMan assesses its weight and decides whether to try to lift it or drag it.

Dr. Young hopes to take this further by dealing with piles of burning tyres. He also wants RoMan to be able to use its body weight in the way a human being might to push open a tight door or to move heavy furniture by bracing against a wall. One problem with RoMan is that it is still impractically slow. It often takes 10-15 seconds to decide what to do. RoMan will also need to learn to deal with a wider range of objects.

However, the device's future could be bright. Beyond military applications, its following generations might work in warehouses, pick fruit, clear litter or tidy people's homes. They might even collect rocks from the surface of Mars.