Weaknesses in Tesla's Software-based Autonomous Driving Technology

Navigant Research’s Sam Abuelsamid comments on Tesla’s approach to self-driving technology in USA Today

According to a USA Today article, Tesla, one of several companies leading the path to fully autonomous driving, may be leaving a key technology out of their self-driving system which is causing consumer safety concerns.

Instead of using a combination of technology including cameras, a light ranging and detection laser system (LiDAR), and radar, like its competitors, Tesla’s semi-autonomous vehicles are only using cameras and radar for its Autopilot driver assistance system. 

Sam Abuelsamid, a senior research analyst with Navigant, offered his own expertise in the article.

"I think you absolutely need all three types (cameras, LiDAR and radar) because each one has strengths and weaknesses," said Abuelsamid.

Abuelsamid explained that while cameras may excel at recognize objects, they are less reliable at seeing objects in complicated lighting conditions. Meanwhile, LiDAR uses lasers to bounce off objects to create images, which can avoid lighting problems caused by weather and darkness.

"Tesla's trying to do it on the cheap," Abuelsamid said. "They're trying to take the cheap approach and focus on software. The problem with software is it's only as good as the data you can feed it."

According to USA Today, Tesla's omission of LiDAR is raising questions surrounding recent crashes by their cars in Autopilot mode. In addition, other experts agree that LiDAR may be overlooked in favor of cost.

“With prices estimated as high as $75,000 per vehicle in recent years, it's not difficult to see why. But those in the industry say prices will come down dramatically in the future,” wrote USA Today.

Read the full article
Back to top