India’s space agency is set for its second attempt at a lunar landing with a spacecraft loaded with extra fuel, designed for myriad failures, and assigned a landing area 40-fold larger than earmarked for its predecessor that crash-landed on the moon four years ago.

The Indian Space Research Organisation (Isro) chairman S. Somanath on Monday provided granular details of how the space agency’s engineers have tried to fail-proof the Chandrayaan-3 spacecraft now sitting atop a rocket at the Sriharikota spaceport awaiting launch at 2.35pm on Friday.

Chandrayaan-3, Isro’s third lunar mission, is expected to demonstrate a soft landing of its lunar module near the lunar south pole on August 23-24. The module has a lander and a rover, both packed with payloads for scientific experiments for one lunar day — 14 Earth days.

“Instead of a success-based design … we’ve opted for a failure-based design,” Somanath said. “We looked at what all can fail and how to protect (against) them. We looked at sensor failure, engine failure, algorithm failure, calculations failure, Whatever may fail, we want it to land at the required speed. Different failure scenarios have been calculated and programmed inside,” he said on the sidelines of a space industry conference.

Isro’s Chandrayaan-1 launched in 2008 was a lunar orbiter that had helped detect signatures of water molecules on the Moon. Chandrayaan-2had made it all the way to the Moon in September 2019, but things went wrong during the final descent, causing the lunar module to hit the Moon at high velocity.

Spacecraft and mission designers have also incorporated changes in the Chandrayaan-3 spacecraft and the mission profile after taking into account the cascade of failures that had hit Chandrayaan-2.

During the Chandrayaan-2 lander’s descent, the engines designed to reduce its velocity developed higher thrust than expected, Somanath said. The increased thrust led to errors that accumulated, prompting the lander to make fast turns, but its turning ability was limited by software.

The earmarked landing area posed another challenge. Isro had asked the spacecraft to land in a 500m by 500m area. As the lander was trying to reach the area, it was also close to the ground. This, Somanath said, led to a contradictory requirement — reaching the area and at the same time, achieving a low velocity.

“This became mathematically difficult to do in the available time,” Somanath said. In a nutshell, he said the spacecraft’s ability to handle deviations from expected performance was “limited”.

For Chandrayaan-3, the landing zone has been expanded to 4.0km by 2.5km, 40 times bigger than for the preceding mission. “Suppose, the performance is poor, it may land anywhere in the area. Second, we’ve added more fuel, so it has more capability to handle any dispersions or move to alternative landing sites,” Somanath said.

Isro’s Geosynchronous Satellite Launch Vehicle Mark 3 will ferry Chandrayaan-3 into space.

The landing next month will use high-resolution images taken by a camera aboard Chandrayaan-2’s orbital module.

“Last time, we took images from the lander, and analysed them on the ground, telling (the lander) how to do the landing. This time, we know the landing site, the boulders, the craters are all well-mapped, programmed into the lander,” Somanath said.

The scientific instruments on the lander and the six-wheeled rover will study the lunar surface, look for signs of lunar seismic activity, and study any ions near the lunar surface.