RADAR, LIDAR and beyond……
To start the New Year, we wanted to take a look at some new and upcoming technologies that will eventually benefit all of us. We have all seen reports of driverless transports as a way of the future but how does that technology affect us today? The reality is that autonomous vehicles will eventually arrive but it will be at least 10 years before they are truly driverless. A recent incidents with Tesla’s auto drive system as well as the Google car being involved in a crash have put a bit of a reality check on some of the autonomous vehicle boosters. We see this as part of the future but it’s also not going to be a quick as some would like.
Autonomous vehicles use a number of sensors of various types to help “see” what is in front of it. – cameras, ultrasonic sensors, RADAR and more recently LIDAR. The first three are more established and more economical to deploy. LIDAR is newer and has not yet achieved an economy of scale. However it is potentially the most beneficial for heavy duty vehicles. LIDAR uses light (typically laser) scanning and ranging to build up a detailed 3D image of the surroundings. These images are of a very high resolution – aircraft mounted units have been able to map terrain at 12-inch resolution from cruising altitudes. Light is at a lower wavelength than the radio waves used in RADAR, and this allows it to see much smaller object because light has better reflectivity than radio waves on the same object. One of the powerful aspects of this technology is that this image can be compared to the data captured by the other sensors. In its current state, however, it is bulky ad cumbersome – it’s part of why the Google or Uber autonomous vehicles have those huge domes on top of them. They are currently complicated mechanically and can cost as much as your average car. Several large automotive suppliers are working on miniaturizing LIDAR units and a recent article in the Economist suggests that these will start appearing in vehicles within three year.
Currently one of the largest manufacturers of RADAR equipment has turned their expertise towards making a small LIDAR unit. Infineon has discovered a way to make RADAR equipment with a standard silicon-based manufacturing process. By doing so they also were able to integrate many functions onto one chip, bringing the price to approximately a tenth of what it was 10 years ago. They are currently one of many companies looking to do the same with LIDAR.
The current “state of the art” sensor is capable of scanning up to 5000 data points per second at a range of 250 feet. Once they go into production they are expected to be units that can be mounted to a windshield and cost around $250 each!
Okay, we are all now probably saying, what’s in this for us?
First of all LIDAR has the ability to “see” materials such as rubber that RADAR has difficulty with. A RADAR sensor is not going to see that tire tread of the other truck that’s sitting in the lane your truck is driving in. A camera system should see it a bit further out and pre-warn the LIDAR unit. RADAR will help see through fog, rain or snow to see the other vehicles around you, and it will combine its data with the others to help determine the speed you are approaching other objects at. Ultrasonic sensors will help for short range tasks like parking assistance.
Practical applications include lane departure avoidance systems, adaptive cruise controls, collision avoidance and eventually platooning. Some of these technologies are already in production vehicles but their cost will start to come down and their reliability will increase with the addition of different sensors.
For example, adaptive cruise control is already available on many heavy duty trucks. It looks at the speed of the traffic the driver is in to make adjustments to the speed the cruise is set for. Many of us have had the experience of heading down the I-75 with the cruise set at 70 while heading south of Toledo. You get a couple of hours of mostly worry-free travel until you hit that (seemingly) never ending stretch of construction that starts in Dayton and makes its way through to Cincinnati. Now it’s back to using the pedal, adjusting your speed constantly and quite honestly you’re getting tired. That’s an experience your drivers face in many locations day in, day out. What if that adaptive cruise could make those adjustments for you? Yes, there is going to be a lower limit where the driver must take over, but wouldn’t this take away some of the stress and fatigue encountered by drivers going through rush hour traffic in cities like Toronto or Atlanta? Similar to an automatic transmission, this sort of technology means there is one less thing for the driver to worry about and additional time for him/her to focus on what the other drivers are doing.
Now combine that adaptive cruise control with a collision avoidance system that is looking out in front of you for objects that may become a hazard. Now you have a vehicle that can work with your driver, seeing some of the things that they may not be focusing on in the large number of inputs they face every minute on the road. What if those forward sensors could not just adjust the speed but also apply the brakes while taking into account that car that’s following too close behind the trailer? Is this going to replace the driver? Probably not. Regardless of how many sensors and how much processing power we put into a truck it still will not match the human brain’s capacity to handle multiple stimuli. What this will do is back the driver up for those times when they need it – rainy conditions in the middle of the night, construction zones with narrowed and merging lanes or that load that needs to get there but the driver is fatigued or sick. Your business will still need to recruit properly qualified drivers but there is no reason we can’t give even a seasoned veteran a little help. You will also want to discuss these systems with your insurance company while you are in the specifications stage as they may offer a discount that will reduce your payback period. At the end of the day you need to make sure that this sort of technology will work with both your current operations and where you want to be 4-5 years from now