You may be familiar with Light Detection and Ranging (LiDAR) sensors from reading about the recent explosion in research and development for self-driving cars. While LiDAR is central to autonomous vehicles, including commercial trucks, it’s also rapidly becoming a key technology in industries as diverse as agriculture, energy, robotics, the military, Unmanned Aerial Vehicles (UAVs), mapping and even virtual reality.
In all these fields, LiDAR represents one of – if not the most important – remote sensing modality. However, there’s an expanding mélange of LiDAR technologies in the marketplace today. This two-part series demonstrates how 360º rotating HD LiDAR sensors – specifically those developed and built by the widely recognized leader in the field, Velodyne LiDAR – embodies the benchmark for the technology and its many uses.
In part one, we’ll discover how Velodyne LiDAR provided the trailblazing technology in early autonomous vehicle development, and how the company’s innovative sensors still set the standard in terms of size, flexibility and pricing. In part two, we’ll delve into how Velodyne LiDAR also has become the de facto standard for a variety of new and established industries due to its innovative and time- and money-saving applications.
Precision Form of “Seeing” for Robots
Taking laser measurements of every object in the immediate vicinity represents a precision form of “seeing” for autonomous vehicles, UAVs, robots, mapping devices and any other industry where automation meets the physical world. Compared to radar, sonar, stereo cameras and night-vision, LiDAR provides the clearest, cleanest and most prolific sensor data of any modality in the marketplace today.
LiDAR is not — or should be — the only remote sensing solution providing ‘sight’ to automated vehicles and other robotic platforms, as this function is too crucial to rely on a single technology. But as a key component in a suite of sensing modalities, LiDAR has earned its place of prominence – and fills in the gaps where other sensors leave off.
Yet LiDAR is not as generic as its increasingly recognized acronym might suggest. Lasers may be somewhat interchangeable commodity parts, but not every LiDAR technology is alike. To the contrary, there are ideal LiDAR solutions, and then there’s everything else.
To give a sense of how LiDAR can best perform in present-day use cases – and how rotating, multi-laser (a.k.a. multi-channel) LiDAR offers the simplest and best solution for most applications – we’ll cover how some of Velodyne’s most successful automotive use cases – from the origins of modern self-driving cars to the latest cutting-edge autonomous vehicles.
First, here’s a brief introduction to Velodyne’s LiDAR technologies.
- The HDL-64E, the gold-standard of 360º high-definition LiDAR that traces back to DARPA’s self-driving Urban Challenge in 2007 and carries forward to today, with widespread contemporary application in autonomous vehicles as well as use in mapping, UAV and robotics.
- The HDL-32E, a smaller, rotating, soda-can-sized hybrid LiDAR that continued the HDL-64E’s tradition of innovation and transformation of industries it enters.
- The VLP-16, also called the Puck™, further miniaturized and modernized the innovations of its predecessors. Velodyne has also developed two related versions, Puck Lite™ for UAVs and other weight-conscious applications in which every payload gram counts, and Puck Hi-Res™, for concentrated-beam LiDAR applications in mapping, transportation and security.
- Velodyne’s latest generation Puck sensor, the VLP-32C Ultra Puck™, doubles the channel count to 32 compared to the Puck and similarly boosts its resolution and data rates. The Ultra Puck has been engineered to fit the technical requirements and form factor constraints of automotive applications such as advanced driver assistance systems (ADAS) and self-driving cars.
Furthermore, in 2017 Velodyne announced an all-solid-state LiDAR breakthrough called Velarray that will reduce pricing, increase performance and – in its extremely portable and minuscule form factor – further broaden the range of applications and industries it will transform. Velarray’s development is ongoing with customer testing expected to start soon.
Overall, Velodyne’s LiDAR sensors are the highest performance LiDAR systems on the market for their size, weight and power class. They are all Class 1 eye safe and engineered to high standards of water and dust ingress protection (IP 67 rating for the HDL-32E). Details and technical specs for each unit, please click here.
“The Clear Winner Was Velodyne”
Long before the current self-driving car frenzy, in 2007 the Defense Advanced Research Projects Agency (DARPA) held its first Urban Challenge – and five of the six early autonomous vehicles that finished the competition used Velodyne LiDAR. Even with world-class robotics, research and automotive teams participating in the event, The San Francisco Chronicle wrote, “In this sensor competition, the clear winner was Velodyne.” Eight years later in 2015, when China held its equivalent of the Grand Challenge, the Intelligent Vehicle Future Showcase (IVFC) in Changshu, 17 out of 20 competitors used Velodyne LiDAR, including the top five finishers.
Throughout the first decade of self-driving car development and included in almost every research fleet today, Velodyne’s high-definition, multi-channel 3D LiDAR has remained the key remote sensing technology for highly-automated vehicles. Velodyne’s HDL-64E and HDL-32E sensors collect between 700,000 and 2.2 million data points per second from the environment, extending their 3D sensing out to 200 meters — far beyond the range of cameras. As LiDAR technology matures and develops into essential sensing technology for the future of transportation, Velodyne is now setting the pace in this rapidly-developing field.
Ramp Up of Production of Next-Generation LiDAR
Leading innovators in self-driving technology such as Baidu and Ford both use Velodyne LiDAR and have invested $150 million in Velodyne’s ramp up to produce next-generation LiDAR to commercial scale. A Velodyne mega-factory in San Jose, California will boost the company’s LiDAR production capacity to as much as one million units in 2018, enabling economies-of-scale that carmakers demand for commercial deployment of LiDAR in mass-manufactured cars, trucks, vans and buses.
As previously noted, with Velarray Velodyne is also moving ahead with the breakthrough development of pure solid-state LiDAR technology. Using compact, high-speed, gallium nitride (GaN) circuits developed in tandem with Efficient Power Conversion (EPC) and targeting a price point in the low hundreds of US dollars at high-volume manufacturing scale, Velodyne sets the stage for a new generation of sensors that will maintain the company’s frontrunner position in the industry in performance, price, reliability and range.
Meanwhile, Velodyne’s existing line of 360º rotating LiDAR sensors — the industry-leading HDL-64E and HDL-32E, the portable VLP-16 Puck and the groundbreaking VLP-32C Ultra Puck — remain the most widely-deployed suite of LiDAR sensors in the transportation industry. For example, in 2015 Frost & Sullivan recognized the VLP-16 with its Product Leadership Award for ADAS automotive sensors.
“The VLP-16 offers tremendous growth potential in the future, with sales expected to cross several thousands of units when autonomous vehicles become a norm in the industry,” the industry analysis group noted in its report. “As vehicles become more and more autonomous, companies that can bridge the gap between cost and quality will become the industry’s ‘preferred vendors.’ Despite the current market being in a nascent stage, 9 out of 10 OEMs use a Velodyne sensor, giving clear evidence of the company’s excellent reputation in the industry.”
Building on the VLP-16’s success, Velodyne’s new VLP-32C Ultra Puck promises 32 LiDAR channels with a range of up to 200 meters and accuracy of +/- 3 cm – yielding up to 1.2 million data points per second (dual return mode). Developed for automotive applications, the Ultra Puck is cost-competitive with rival LiDAR technology.
The Ideal Solution
Across nearly all sensing challenges, multi-channel rotating LiDAR like Velodyne’s remain the ideal solution compared to other sensing modalities like short-range and long-range radar, sonar, stereo cameras and infrared. In all challenging and diverse operating environments – near-field and far-field sensing, narrow vertical field-of-view and wide-angle, trajectory tracking, night operation, high-contrast and low-contrast ambient lighting – LiDAR stands as either the undisputed best or one of the top sensing solutions on the market.
However, under any modality rain and snow are one of the most challenging environments to master for ADAS and self-driving technology. Yet, even with inclement weather, progress has been reported.
Using one of its 32-channel Velodyne units, the HDL-32E, Ford has developed software that greatly improves its performance in snow and rain. Ford’s project “Snowtonomy” integrates both first reflection and subsequent reflections from each of the LiDAR unit’s lasers. The later reflections (from, say, refracted laser light through a raindrop) typically bounce off the ground. Using algorithms, Ford can tease out reflections from snowflakes or raindrops and simply subtract them.
A Transformative Technology
It’s no exaggeration to say that LiDAR is a transformative technology in the increasing number of industries it enters – not just automotive. Whether the LiDAR unit is mounted on a car, commercial truck, UAV, helicopter, boat, robot or virtual-reality camera, its 360º high-definition 3D “seeing” capability makes possible technologies that might otherwise seem impossible.
The type and quality of LiDAR sensors in the marketplace are as varied as the fields they’re revolutionizing. And while the race for second- or third-rate technology may be heating up, for best-in-class LiDAR there’s no competition with multi-laser, rotating, hybrid solid-state sensor. Velodyne LiDAR invented the category at its founding in 2007 and remains the dominant player in the field to this day.
To learn more about Velodyne LiDAR technology, visit Velodyne LiDAR’s website and contact a sales representative for a free trial today.
About Velodyne LiDAR
Founded in 1983 and based in Silicon Valley, Velodyne LiDAR, Inc. is a technology company known worldwide for its real-time LiDAR sensors. The company evolved after founder/inventor David Hall developed the HDL-64 Solid-State Hybrid LiDAR sensor in 2005. Since then, Velodyne LiDAR has emerged as the leading developer, manufacturer, and supplier of 3D real-time perception systems used in a variety of commercial applications including autonomous vehicles, vehicle safety systems, 3D mobile mapping, 3D aerial mapping, and security. Its compact, lightweight HDL-32E sensor is available for many applications including UAVs, while the VLP-16 LiDAR Puck is a 16-channel real-time LiDAR sensor that is both substantially smaller and dramatically less expensive than previous generation sensors. Finally, the VLP-32C Ultra Puck™ offers the long-range, high-resolution and surround view required for the world’s most advanced autonomous vehicles.