So when will self-driving cars really come to a driveway near you? The answer depends in part on whether such cars require a type of sensor called lidar, short for “light detection and ranging.” Most groups developing autonomous vehicles see chip lidar as a critical part of the sensor suite required for safe operation, because it allows a detailed 3D map of the vehicle’s environment to be constructed with much more fidelity than can be done with cameras.
Elon Musk, though, has been pushing Tesla to adopt a controversial cameras-only approach to autonomous driving. “Humans drive with eyes & biological neural nets, so makes sense that cameras & silicon neural nets are only way to achieve generalized solution to self-driving,” Musk tweeted in 2021. The mechanical complexity and high cost of most lidar sensors—which not long ago would have added tens of thousands of dollars to the price of each vehicle—no doubt helped shaped Musk’s views. As early as 2016, he declared that “all Tesla vehicles exiting the factory have hardware necessary for Level 5 autonomy”—meaning that cars with cameras and computers alone have what’s needed for fully autonomous driving.
Seven years and many crashes later, Tesla has not progressed past Level 2 Autonomy, and traffic-safety specialists are questioning Musk’s rejection of lidar. Requiring pricey sensors, though, would slow the widespread rollout of both advanced driver-assistance systems and fully autonomous driving. But reducing the cost of these sensors to a level that would satisfy automakers has remained an elusive goal for lidar manufacturers, which must also consider how to add their devices to cars without detracting from vehicle aesthetics.
We and others at our company, Analog Photonics, which spun out of MIT in 2016, hope to break this impasse. We are developing a tiny, chip-scale phased-array lidar that promises to slash costs and simplify integration. Here we’d like to explain some of the technical challenges we’ve encountered and how very close we are to commercialization.
From Radar to Lidar
Today, more than half of new cars are equipped with one or more radar sensors. These sensors are solid state, cost manufacturers less than US $100 each, and are small enough to be inconspicuously placed around the vehicle. They are used for a variety of things, including automatic emergency braking and adaptive cruise control, as well as lane keeping and other advanced driver-assistance functions.
But this wasn’t always the case. Early automotive radars were large, mechanically steered, emitted short pulses of radio waves, and had limited performance. But the move to electronic scanning and continuous-wave emissions in automotive radars brought performance advancements and cost reductions, which in turn ushered in their widespread use.
Lidar is now undergoing this same evolution.
For the complete article CLICK HERE.
Note – If you liked this post click here to stay informed of all of the 3D laser scanning, geomatics, UAS, autonomous vehicle, Lidar News and more. If you have an informative 3D video that you would like us to promote, please forward to editor@lidarnews.com and if you would like to join the Younger Geospatial Professional movement click here
