Drone?machine learning duo speeds up meteorite hunt無人機和機器學習加速隕石尋找

羅興濤

隕石可以提供關于早期太陽系是什么樣的的線索，但是找到它們是相當困難的。現在，一些科學家正在轉向無人機和機器學習，以更加高效地幫助發現新落地的隕石。

主題語境：天文研究 篇幅：354詞 建議用時：7分鐘

1 Meteorites （隕石） can offer clues to what theearly solar system was like. But finding them israther difficult. Now， some scientists areturning to drones （無人機） and machine learningto help spot freshly fallen meteorites muchmore efficiently. “A team of six people on ameteorite?hunting expedition can search about 200，000 square meters per day，” saysSeamus Anderson， a planetary scientist in Australia.

2 Around 2016， Anderson began adopting the concept of using drones to take pictures ofthe ground to look for meteorites. That idea blossomed into a PhD project. In 2022， heand his colleagues reported their first successful recovery of a meteorite spotted with adrone. They＇ve since found four more meteorites at a different site. Drone?basedsearches are much faster than the standard search way.“ You＇re going from about 300days of human effort down to about a dozen or so，” he says.

3 Anderson and his workmates have used drones to search for meteorites in remote partsof Western Australia and South Australia. The team is tipped off about a fall site by networks of ground?based cameras that track meteoroids flashing through the Earth＇s atmosphere.The researchers have to do a series of fun but difficult work before the hunt.They pack a four?wheel drive vehicle with drone and computer equipment， batterycharging stations， generators， fuel， food， camping equipment， tables， chairs and muchmore. The drive to the fall site can take more than a day， often on rough or non?existentroads. Anderson says“， You hope you don＇t pop a tire.”

4 After arriving， the team flies its primary drone at an altitude of about 20 meters. Its cameratakes an image of the ground once every second， and the scientists download thedata every 40 minutes or so when the drone lands to receive fresh batteries. A typicalday of flying can net over 10，000 images， which are then divided digitally into 100 millionor so smaller sections. Those“ tiles”， each 2 meters on a side， are fed into a machinelearning algorithm （算法） that has been trained to recognize meteorites based onimages of real land rocks which are spray?painted black.

Reading Check

Detail

1. Why do the scientists study meteorites？

A. To spot the planetary course.

B. To promote machine learning.

C. To test the functions of drones.

D. To explore the past of the solar system.

Inference

2. What does Anderson say about drone?based searches in paragraph 2？

A. Their barriers. B. Their causes.

C. Their efficiency. D. Their concept.

Detail

3. Which words can best describe the preparations of the drone search for

meteorites？

A. Fun and light.

B. Smooth and flexible.

C. Difficult and unpleasant.

D. Interesting and tough.

Detail

4. Why is the machine learning algorithm used according to the last

paragraph？

A. To analyze drone data.

B. To train drone recognition.

C. To divide images into sections.

D. To receive fresh batteries.

Language Study

Ⅰ. Difficult sentence in the text

Those“ tiles”， each 2 meters on a side， are fed into a machine learning algorithm thathas been trained to recognize meteorites based on images of real land rocks which arespray?painted black. 那些邊長為2米的“瓦片”被輸入到一個機器學習算法中，該算法經過訓練，能夠根據噴涂成黑色的真實的陸地巖石圖像識別隕石。

【點石成金】本句是一個復合句。each 2 meters on a side為獨立主格結構，對名詞“tiles”進行補充說明；that引導的是一個定語從句，先行詞為a machine learning algorithm；based on images of real land rocks為過去分詞短語作方式狀語；which引導的是一個定語從句，先行詞為real land rocks。

Ⅱ. Text?centered chunks

take pictures of 拍攝……的照片

look for 尋找

blossom into 發展成

search for 搜尋；尋找

a series of 一系列……

at an altitude of 在……的高度

divide...into 把……分成

be based on 以……為基礎