New MEMS beam steering technology is expected to significantly reduce the cost of lidar

This micro beam steering device based on MEMS technology is expected to achieve lighter and lower-cost lidar

LiDAR is one of the key technologies for self-driving cars to detect and identify surrounding objects. According to MEMS Consulting, a research team at KTH Royal Institute of Technology in Sweden has been committed to studying the beam steering technology at the core of LiDAR systems for a long time and has developed a beam steering device that is more economical, smaller and more resource-efficient than previous technical solutions.

Carlos Errando-Herranz, a postdoc at KTH's Institute of Micro and Nano Systems, said: "With our technical solution, the cost of this lidar can be significantly reduced to about $10 after mass production, the weight is only a few grams (including peripheral components), and the power consumption can be as low as hundreds of milliwatts. The research results have been published in the journal Optics Letters.

Beam steering technology has a wide range of applications, such as high-speed optical communications, LiDAR, and medical imaging. Conventional LiDAR beam steering systems use electric motors to deflect mirrors and scan laser beams over a specific area. Such systems are usually large in size and weight, consume high power, and cost thousands of dollars. As a result, such conventional beam steering systems cannot be applied to battery-powered robots, smartphones, drones, in-vivo optical coherence tomography (OCT) probes, and miniaturized, low-cost space division multiplexing (SDM) systems.

近年来，通过利用MEMS微镜和光栅缩小了光束操纵系统的尺寸，从而显著降低了成本和重量。然而，这些激光雷达系统的组件（如激光器、扫描装置、探测器及其它电子器件）仍然是独立制造的，并且组装成本较高。因此，进一步的多组件集成小型化有潜力以低成本提供更小、更轻、功耗更低的激光雷达系统。

Integrated photonics, especially silicon photonics, can address these challenges through high-density integration of electrical processing and control, beam steering and optical signal processing devices, light sources and detectors. This makes integrated photonic systems superior to free-space optical systems not only in size and weight, but also in cost, integration density and robustness.

Beam manipulation schemes in integrated photonics have mainly focused on optical phased arrays. Optical phased arrays consist of an array of emitters (usually grating couplers) such that the far-field diffraction pattern is highly dependent on the relative phase of the emitted waves. The angle of the output beam is adjusted by tuning the relative phase of these waves using waveguide phase shifters. Such systems can very tightly control the shape and direction of the beam, and previous research work has demonstrated 1D beam manipulation, 2D manipulation with ultra-high angular beam resolution, and lidar measurements. However, commonly used thermo-optical phase shifters have a very important drawback: they consume very high power.

According to Carlos Errando-Herranz, his research team successfully demonstrated low-power beam steering technology in experiments for the first time using MEMS tunable waveguide gratings. The results showed that at a wavelength of 1550 nm, the beam can be steered by 5.6° with a drive voltage of less than 20 V and a static power consumption of less than uW.

This beam manipulation device is based on using a MEMS actuator to change the spacing between the comb teeth of a waveguide grating coupler. Figure 1 (a) shows a schematic diagram of the device. The KTH researchers designed a suspended grating that forms a folded spring, one end of which is connected to a tapered waveguide for optical coupling and the other end is connected to a MEMS comb drive actuator. Horizontal actuation of the comb drive actuator stretches the suspended grating, changing the distance between the grating teeth, resulting in a change in the out-of-plane angular emission of the grating. Figure 1 (b) is a scanning electron microscope (SEM) image of the MEMS tunable grating.

Figure 1 (a) shows the working principle of this MEMS tunable grating before and after actuation. (b) Scanning electron microscope (SEM) image of this MEMS device. The grating acts as part of a soft spring, which is stretched by the comb drive actuator to change the spacing between the grating teeth. (c) Simulation results (color: emitted light intensity) superimposed with analytical estimates (white line) of the effect of increasing the grating tooth spacing on beam steering. (d) Analytical actuation estimates of the comb drive actuator.

“We use the same MEMS manufacturing process as smartphone accelerometers and gyroscopes,” he said, “which means the cost of mass production can be very low.”

Errando-Herranz said the technology could enable more robots or drones to operate or fly autonomously.

"This technology could enable drones to be flown without remote control, for example to deliver emergency medical equipment such as defibrillators," said Kristinn B. Gylfason, associate professor at KTH.

“Robotics and drones are definitely possible applications,” Gylfason said. “Current lidar systems are too expensive for autonomous vehicles, and the automotive industry is very cost-sensitive. Other possible applications include 3D facial recognition in smartphones, such as Apple’s Face ID.”

The uniqueness of KTH's lidar solution lies in the fact that it uses a new MEMS beam steering technology, but it is different from MEMS micromirrors.

“Traditional mechanical lidars are based on an array of lasers mounted on a rotating tower, such as the ‘Family Bucket’ and ‘Super Puck’ from lidar leader Velodyne,” Gylfason said. “Our lidar approach is based on integrated micro-optical mechanics, where we build a tunable grating on the surface of a silicon chip. By changing the grating period, we can determine the direction in which the beam is scanned.”

Compared to free-space optics, KTH's solution is orders of magnitude smaller, lighter, less expensive, less susceptible to mechanical noise, and requires very limited assembly requirements. Integrated thermo-optic phased array systems consume at least five orders of magnitude more power than KTH's device (KTH researchers estimate about seven orders of magnitude due to measurement limitations) and suffer from thermal crosstalk issues, which KTH's solution inherently avoids.

Compared to electro-optical tuning techniques, the power consumption of KTH's device is at least one order of magnitude lower (and more likely three orders of magnitude lower). In addition, the beam steering achieved by KTH is twice as large as the previously reported thermo-optically tunable single grating, and has the potential for even greater angular tuning as the design of future MEMS devices is improved.

The beam-steering technology developed by KTH researchers could offer the low cost and power consumption that industry has long sought to extend artificial vision to battery-powered devices such as smartphones or drones, active probes for in vivo medical imaging, and to increase the bandwidth of optical communications through space division multiplexing (SDM).