Relinquishing human control of the car on a mass scale is, for sure, a major step to take. There are emotional/human acceptance as well as technical barriers to overcome. Succeeding in the latter needs to be based on a huge amount of data, proving and validation. Accumulating all of this through physical, ‘on the highway’ testing alone will take a long time and likely slow the realization of the widespread adoption of autonomous cars and other vehicles. Interestingly, large amounts of proving of the effectiveness and safety of autonomous driving technology, with the right publicity, should also help address the human apprehension element – especially for those with many years and a love of ‘hands-on’ driving.
The time it takes to gather real-world data from autonomous vehicles on the road is why tools such as NVIDIA’s DRIVE™ Constellation – an open, cloud-based platform that can perform bit-accurate simulation for large-scale, hardware-in-the-loop testing and validation of autonomous vehicles – are so important. The viability and success of such platforms in supplementing and corroborating real-world testing for the proven, safe and earlier realization of driverless cars, is dependent upon collaboration with specialists in specific technologies that are cornerstones to the implementation of autonomous vehicles.
ON Semiconductor, with its leading image sensor knowhow and technologies, is one such company. Its recent announcement of an agreement to work with NVIDIA underlines its reputation and will see the company’s CMOS image sensor modelling technology being used as a primary function to simulating real-world sensor performance within the DRIVE Constellation.
The image sensor model receives both scene information and control signals from the DRIVE Constellation to calculate and output a real-time image based on the inputs. It then transmits the simulated image back to the DRIVE Constellation for processing. The complex sensor model will utilize all critical parameters in the path to provide an accurate output of a real-world image sensor.
The changes to our everyday lives could be massive, as too will be the effect on the safety and environmental impacts of getting from A to B for millions across the globe. The ‘virtual proving’ provided by tools such as NVIDIA’s DRIVE Constellation and supported by collaborators such as ON Semiconductor, will benefit those engaged in the ground-breaking work to take autonomous driving from concepts under test to mainstream reality.
