Optics, a small word, yet a powerful game changer in the world of LiDAR.
There are several elements in any machine or device that makes it work. People generally have this misconception that the laser, detector or signal processing unit is the most important component of a LiDAR… but it all boils down to the optics. Now that we have cleared up one myth, the fact remains: optics are the powerhouse of any LiDAR system. They are the American muscle cars of the LiDAR world- if you get the optics right, it provides the torque - that special superpower that enables a LiDAR.
Simply put, the bigger the optics in a LiDAR system, the better the performance. However, the same cannot be said for amplification, when amplifying electrical signals - you will also be amplifying the noise. Optics on the other hand collect more signal without collecting more noise. A major win in any complex system!
When it comes to optics, there are two key questions: firstly, how will you project the beam as far as possible, without compromising eye safety? And secondly, how can you collect as many photons as possible? Sound easy? Not quite!
You need to create the most uniform, outgoing beam with no optical distortion. Obtaining beam uniformity, technically known as ideal collimation (the degree of parallelism of a beam) is difficult. For example, if using a semi-conductor laser, a distorted ellipse pattern is produced. Now, you need to transform this beam pattern into a perfectly uniform parallel beam. In most cases, it requires multiple optical elements to get this collimation to be ideal and efficient.
Collecting photons is the primary objective, it’s all about getting the photons down to a detector and focusing those photons to an exact spot that is the size of the detector. The optical system can now be treated as an energy gathering device and not an imaging device. Ensuring that your lenses are optimally sized to guarantee clear imaging as well as efficient collection of photons involves something of a dance between the various parameters of engineering, because you don’t want to “overtorque” the engine at the expense of the gearbox – in other words, lose out on one area because you’ve enhanced the other.
Something to ponder on: What’s the point in securing an enviable collection of photons if you get a lot of thermal distortion?
Still on the topic of lenses - a larger lens collects more light than a smaller one. Below are some key considerations when thinking about lenses: Choose your material with care - glass may seem like the best choice however remember that the devices that make use of LiDAR systems, such as drones, may be prone to crashes. Glass tends to fracture and scatter into small pieces on impact, whereas acrylic is more ductile, and so fractures in a gentler way. Purists may argue that acrylic doesn’t give an identical performance to glass, and while this may be significant in a laboratory setting, in the real world, the discrepancy between the two is so minimal that it becomes irrelevant. Coatings on lenses are used to improve scratch resistance as well as avoid concerns such as backscatter.
Finally, what about configuration? This is critical because you want to avoid backscatter; the ghost return signal which comes about if your lenses scatter light which is picked up by the detector. While it may help to focus only on the center of the light returned, and ignore the rest of the pattern, this wastes an enormous amount of energy, affecting eye safety. A better bet is to isolate lenses so that they are unable to ‘see’ each other. Cylindrical lenses can also be configured so that the focal length of one addresses the fast axis, while the other deals with the slow, wider axis. This approach helps secure the best collimated beam.
Here comes the complicated part of the equation : the receiver. A larger lens on the receiver captures more photons. Bigger systems however, are more unstable! So, what can be done? The efficiency of the lens needs to be improved; this can be achieved by minimizing spherical aberration by using doublets (pairing lenses with different curvatures, refractive indices and amounts of dispersion). This method achieves the sharpest focal point.
To sum it up, it’s all about collecting photons in the most efficient way possible.