Chandrayaan-3: How AI helped India create history by landing on the moon

India on Wednesday became the first nation to reach the moon’s south pole as the ambitious Chandrayaan-3 spacecraft landed successfully on the surface of the earth’s satellite per the schedule. After a landing sequence that lasted about 15 minutes, Vikram Lander touched the moon’s surface, putting the country on the map of countries that are now present on the moon. At play were several technologies that made the mission successful, but Artificial Intelligence (AI) played a crucial role in the wee hours of the landing. The descent of the lander was entirely controlled using AI, making it one of the latest achievements in the field.

The descent, which began at 5.45 pm on August 23, Wednesday, was managed by AI software that used instruments like velocimeters and altimeters to calculate the speed and position of the lander above the surface of the moon. The AI-enabled software used the cameras on the orbiter and the lander to detect potential landing sites. These cameras are the Lander Position Detection Camera (LPDC) and the Lander Hazard Detection & Avoidance Camera (LHDAC).

Chandrayaan-3 Mission:
Updates:

READ MORE

After Chandrayaan 3, these are the next missions ISRO is working on

The communication link is established between the Ch-3 Lander and MOX-ISTRAC, Bengaluru.

Here are the images from the Lander Horizontal Velocity Camera taken during the descent. #Chandrayaan_3#Ch3 pic.twitter.com/ctjpxZmbom

— ISRO (@isro) August 23, 2023

A part of the descent involved changing the orientation of the Vikram Lander from horizontal to vertical for a smooth landing. According to ISRO chairman S Somnath, this was one of the most complicated stages of the landing sequence. India’s last moon lander Chandrayaan-2 ran into some problems, which is why the space agency took the help of AI to ensure the perfect switch to the orientation. A dedicated team trained the AI software through various simulations that allowed it to change Vikram Lander’s orientation from horizontal to vertical during the last stages of the descent. With the help of the Lander Horizontal Velocity Camera (LHVC), AI managed to change the orientation of the lander by tilting it at 90 degrees.

Using the learnings from the last attempt, ISRO was wise enough not to rely entirely on AI and made the lander foolproof by all means. The spacecraft features additional sensors to ensure safety even if AI fails to sense the surroundings to prompt the orientation switch.