It sounds like Volvo have missed a trick if they are using software NN to interpret reflected lidar pulses in EVs. it's not so important with ICEs because refilling is not a problem, but EV batteries take a long time to recharge.Yeah I've never been a big fan of running ADAS on GPU's.
"I think that the foveated lidar will probably allow autonomous driving in excess of 100 kph because it will enable the receiver to detect distant objects with greater certainty than vanilla lidar"- Correct. MB might be pivoting to do a trade-off. Take the power hit for the extra lidar range for now to get approval to go over 100 km/h
MB can take the variable pulse rate lidar pulse transmitter from Luminar and use Akida in the receiver because Akida can process the reflected signals more efficiently than a software NN.
In the EV business kilometers count. It is a major selling point, so an extra 30 km per charge is important.
Combining Akida with the Luminar variable pulse rate laser may be MB's secret sauce.