Lidar is now doing what happened to Radar some years back. Impressive to think that this is happening at near optical frequencies.

Seems like the difference between the original spark gap transmitters and more sophisticated signal modulation methods.

Thank you for the explanation. By the way, is frequency modulation means that you change the wavelength 1550nm to 1550nm +/- delta_lambda? Or is that means you modulate 1550nm light's amplitude to a certain frequency?

@silctechnologiesinc.5567

3 жыл бұрын

Frequency modulation means that you change the wavelength 1550nm to 1550nm +/- delta_lambda.

Passive is the lock, and quantum detection is the key. You guys should work on that.

But you didnt mention the limited FoV of FMCW Lidar as compare to the ToF Lidar..

Ok let's say if the target is moving in subsonic speed ??? Can theses lidars could replace the pulse Doppler radar ???

Is it true in case of radar also ?! Please answer

@silctechnologiesinc.5567

3 жыл бұрын

Radar operates at mm-Wave wavelengths (e.g. 77GHz). These wavelengths are about 1,000 times larger than the optical wavelengths with different reflectance characteristics. Hence, radar doesn't provide the angular resolution needed to resolve smaller objects and a coke may have a much stronger reflection than a human next to it, for example.

@ruchirdave856

3 жыл бұрын

@@silctechnologiesinc.5567 ok got u, actually I was looking for difference between tdr and fmcw radar level transmitter and ur videos got me a clear scene, thanks a lot

For autonomous cars, which one is better FMCW or TOF?

@silctechnologiesinc.5567

2 жыл бұрын

FMCW is by many considered to be the next generation LiDAR. Perception software based on ToF LiDAR requires a lot more machine learning and training compared to FMCW LiDAR. Here is a quick example: If you scan with ToF LiDAR and you get a return let's say at 200m, there is still no certainty that there is an object or if it's just noise. The algorithms would have to make an assumption that there is an object and track that assumption through multiple frames to confirm there is an object. Once confirmed, the algorithm would determine the motion vector of the object and after that predict where the object is going to be in the future. In comparison, with an FMCW LiDAR, you might get a return at 200m and an associated motion vector. Let's say the object is moving at 50mph towards you. Right there, the perception algorithm would already have a pretty good idea that there is an object, what the object is, and where it's going to be next. All with very low latency, compute power, and without training.