Computer vision and advanced imaging technologies have given rise to a number of revolutionary technologies—one of the most significant being advanced driver assistance systems (ADAS) and autonomous vehicles.
Whether the vehicle’s imaging comes from a camera, radar, or LiDAR, one constant is that the sensors need unobstructed vision to do their job. For this reason, the automotive industry has sought ways to prevent and remove the buildup of debris and contaminants on vehicle image sensors.
This week, Texas Instruments released two new complementary chips that work in unison to keep image sensors contaminant-free. In this article, we’ll look at the company’s proposed lens-clearing method and the new chipset released to support it.
In modern vehicles, external-facing image sensors are exposed to the outside environment, where dirt, debris, and water can land on a sensor’s lens. This can obscure the sensor's line of sight and compromise the entire system’s functionality. Cleaning the lens is a mostly manual process and results in unwanted device downtime.
While there are several proposed techniques for automated lens cleaning, Texas Instruments supports ultrasonic lens cleaning: the process of vibrating an image sensor’s lens to remove contaminants.
The theory behind ultrasonic lens cleaning is that every material has a resonance frequency, which is the frequency at which the material will naturally vibrate. The goal of ultrasonic lens cleaning is to force the lens to vibrate at its resonance frequency when contaminants are present, causing any contaminants on the surface to be vibrated off and out of the sensor’s field of view.
In practice, this can be implemented by piezoelectric transducers, devices that change shape when a voltage is applied to them. If the applied voltage is alternating in nature (in other words, AC), then the material will resonate at the frequency of the AC signal. Hence, if an AC signal is applied to the transducer with the same frequency as a lens’ resonance frequency, the transducer can help remove contaminants from the lens.
This week, Texas Instruments released two new complimentary chips to enable ultrasonic lens cleaning in automotive solutions. The two chips, the ULC1001 and the DRV2901, are meant to work in unison in a piezoelectric transducer system to help detect and remove contaminants from an image sensor in real-time.
The ULC1001 is a configurable PWM modulator with current- and voltage-sensing capabilities designed specifically for piezo-based lens cleaning systems. The controller is integrated with on-chip algorithms to automatically sense and clean contaminants with lens system calibration. The DRV2901, on the other hand, is a lens cleaner transducer driver. The chip features a one-channel half-bridge that outputs a modulated control signal, requiring only a simple passive LC demodulation filter on the output to deliver high-quality, high-efficiency amplification to a piezoelectric transducer.
Together, these two chips work to sense the state of the piezoelectric transducer, detect the presence of contaminants through integrated DSP algorithms, and eventually control the transducer as necessary to achieve a clean system.