1. What is ToF
ToF stands for time of flight. It is a ranging method that measures the time it takes for a signal such as ultrasound, microwave, or light to travel from an emitter to a reflector and back. Sensors that implement this measurement method are called ToF sensors. Common implementations use infrared or laser light for ranging.
Data produced by ToF sensors provide depth information. When combined with a camera, ToF data can generate 3D images for applications such as pedestrian detection, face-based user authentication, environment mapping with SLAM algorithms, and more. ToF sensors are used in smartphones, robotics, automotive systems, and augmented reality.
2. ToF principle
ToF sensors emit infrared light or laser pulses from a small emitter. The emitted light reflects off objects and returns to the sensor. By measuring the time difference between emission and reception, the sensor determines the "flight time" of the light. Because the speed of light is known and constant, distance is calculated using the formula: (speed of light × flight time) / 2. Practical ToF sensors often use an array format to measure distances across an object surface.
3. Advantages and disadvantages of ToF
The main advantages and disadvantages of ToF sensors are summarized below.
| Advantages of ToF sensors | Disadvantages of ToF sensors |
|---|---|
| 1. Accurate and fast measurements. ToF sensors can detect objects quickly and accurately, and are relatively insensitive to humidity, air pressure, and temperature, making them suitable for indoor and outdoor use. | 1. Scattered light. Very bright surfaces close to the ToF sensor can scatter too much light into the receiver, causing artifacts and unwanted reflections. |
| 2. Long measurement range. ToF sensors are flexible and can detect objects of various shapes and sizes at both near and far distances. | 2. Multiple reflections. In corners or recessed areas, light may reflect multiple times, causing measurement distortion. |
| 3. Safety. Many ToF sensors use low-power infrared lasers driven with modulated pulses. They can meet Class 1 laser safety standards to ensure eye safety. | 3. Ambient light. Strong sunlight can saturate sensor pixels outdoors, making it difficult for the sensor to detect reflected signal from target objects. |
| 4. Lower cost. Compared with other 3D depth-sensing techniques such as structured light or scanning lidar, ToF sensors are generally less expensive. |
Despite limitations, particularly under strong outdoor illumination, ToF sensors are widely used in smartphone imaging, 3D mapping, industrial automation, obstacle detection, autonomous driving, agriculture, robotics, indoor navigation, gesture recognition, object scanning, measurement, surveillance, and augmented reality.
4. ToF sensor chips
Many manufacturers produce ToF sensor chips, including Sony, Infineon, ON Semiconductor, TI, Panasonic, STMicroelectronics, and Melexis. Below is an example description of Melexis MLX75027, an automotive-grade ToF sensor.
MLX75027 features
- VGA (640×480) resolution. Current ToF sensors often have a maximum resolution of VGA.
- Support for 850 nm and 940 nm wavelengths.
- Up to 135 distance frames per second.
- Integrated light source control with modulation frequency up to 100 MHz.
- MIPI CSI-2 serial camera interface using DPHY 2/4 lane.
- Integrated temperature sensor.
Structurally, a ToF sensor resembles a typical camera sensor. Both use a photosensitive array to receive light signals and internal circuitry to convert them into digital image signals. Control is typically via I2C, output is usually MIPI CSI-2, and many devices include an internal PLL.
Other vendors' ToF sensors follow similar principles. For example, some ToF implementations use a single infrared light source to directly measure per-pixel depth and amplitude: the camera module emits modulated infrared light into the scene, the ToF imager captures the reflected light, and depth information is derived by measuring the amplitude and phase difference between emitted and received light.
Conclusion
Although the ToF principle is simple, its applications are wide and increasingly important, especially in AR and autonomous driving. Other depth-sensing methods exist, such as stereo vision and structured light, but ToF imaging offers lower cost and simpler systems, which are distinct advantages. Understanding the technical ideas behind these systems is valuable for solving practical problems.
