It is no secret that self-driving cars have gone from an idea to an arms race in recent years. All over the globe, companies are competing to win market-share in a growing multi-trillion dollar industry. Almost every self-driving car in existence today utilizes one or more LiDAR sensors to help build an accurate picture of its surroundings. These sensors, which have been in use since the 1960’s, have traditionally played a major role in space travel, robotics, surveying, industrial safety, mining, and many other industries.
This article is written to give industry professionals a broad overview of the theory, strengths, and limitations of LiDAR so that they can determine when LiDAR is a good choice for an engineered system. We refer to “robots” in this article. In addition to what people normally think of as robots, our definition encompasses self-driving-cars, autonomous mining vehicles, spacecraft, and any other mobile platform which can drive itself.
LiDAR (Light Detection And Ranging) uses a series of laser pulses in a scanning pattern to generate a “point cloud” of the sensor’s environment. An example point cloud is shown in figure 3. Scanning LiDAR systems constantly send out LASER pulses and see how long it takes them to bounce off of objects and return to the sensor. The time it takes for a LASER pulse to hit an object and bounce back to the sensor is scaled to give the distance to a small part of that object.
Combining these distance measurements in a common coordinate system generates a point cloud. Point clouds can be used to supplement or create maps of all of the opaque objects in a sensor’s field of vision. Scanning LiDAR systems on robots detect objects, build 3D models, and recognize known features on a map in order to deduce the machine’s precise location.
Scanning LiDAR sensors typically have a range of 30 to 120 meters. The directional nature of LASERs makes LiDAR reliable in outdoor environments- as opposed to many other optical depth sensors which get overpowered by the brightness of the sun. High-end units, such as the Velodyne HDL-64E pictured in figure 1, can find the distance to the pavement at shallow angles. This makes them invaluable on self-driving cars and trucks.
One downside of LiDAR that some of our clients have come up against is that it cannot reliably detect optically transparent materials- such as plate glass. LiDAR should, therefore, be supplemented with other sensors such as RADAR and/or SONAR in environments where optically transparent objects are present. As illustrated in fig 3, LiDAR cannot detect colors and has low resolutions compared to cameras.
Scanning LiDAR sensors are relatively expensive- costing anywhere from $1,000 to $100,000. The automotive industry is willing to pay that cost on experimental vehicles because the technology’s many advantages offer safety and supercharge their efforts. Some companies- such as Tesla- have chosen not to rely on LiDAR for their self-driving car technology. LiDAR sensors are expected to go down in price as the automotive industry adopts the technology on a wider scale.
Solid State LiDAR, still in its infancy at the time of this article, does away with moving parts used by scanning LiDAR (fig 2). This decreases the cost of the sensor, reduces the risk of mechanical failure, and allows the sensor to see its full field of view all of the time, decreasing latency.
Solid state LiDAR sensors have been available for several years. As with any new technology, the earliest units on the market have limited resolutions and ranges and are currently rather expensive for the amount of value they provide. That said, the technology is projected to rapidly advance and eventually take the place of its mechanical predecessor.
In mid-2017, Velodyne announced development and testing of a small solid state LiDAR system, dubbed Velarray (Fig 4). The system has a compact package of 125mm x 50mm x 55mm, and provides a 120 degree horizontal and 35-degree vertical field of view. Velarray is not available for purchase at the time of this article, but we will hopefully see these units on the market soon.