2 ms·

I don't mean this to be insulting, but if the intern is taking the point position on "speed" and "vision" you can hardly call that a serious engineering project

I don't mean this to be insulting, but if the intern is taking the point position on "speed" and "vision" you can hardly call that a serious engineering project. There are in fact experts on topics like computer vision, even 20 years ago. They are expensive though.

sokoloff 8y ago
Implement/Operate point of certain subsystems, not theoretical or leadership point. (I take no offense to your comment as it's spot on if that were the case. My apologies if I misled.) I was implementing an improvement and operating our vision system (doubling frame rate and resolution using an existing algorithm across a scaled up processing system). Because of where our code sat in the overall architecture, that I had control systems coursework under my belt, and that I was "done" with my vision processing project much faster than expected, I took on coordinating signals from a variety of other subsystems [radar, lane keeping, sign reading, navigation, etc], computing an overall speed and lateral acceleration target, steer angle, etc, and emitting those and a series of additional signals to downstream systems that controlled the servos and vehicle. Ernst Dickmanns was my boss's project director. I was in no way a "point" on the overall project, which was by all accounts a serious engineering effort. This was part of the Prometheus project (~30-35 years ago, not 20): [0] - https://en.wikipedia.org/wiki/Eureka_Prometheus_Project https://en.wikipedia.org/wiki/Eureka_Prometheus_Project [1] - The vehicle: https://driving.ca/mercedes-benz/auto-news/news/three-decades-ago-mercedes-benz-started-work-on-the-first-autonomous-car https://driving.ca/mercedes-benz/auto-news/news/three-decade...
- erikpukinskis 8y ago
  I'm just saying, there are teams meant to give it a shot, and there are teams meant to provide certainty a thing will get done.

sokoloff 8y ago

Implement/Operate point of certain subsystems, not theoretical or leadership point. (I take no offense to your comment as it's spot on if that were the case. My apologies if I misled.) I was implementing an improvement and operating our vision system (doubling frame rate and resolution using an existing algorithm across a scaled up processing system). Because of where our code sat in the overall architecture, that I had control systems coursework under my belt, and that I was "done" with my vision processing project much faster than expected, I took on coordinating signals from a variety of other subsystems [radar, lane keeping, sign reading, navigation, etc], computing an overall speed and lateral acceleration target, steer angle, etc, and emitting those and a series of additional signals to downstream systems that controlled the servos and vehicle. Ernst Dickmanns was my boss's project director. I was in no way a "point" on the overall project, which was by all accounts a serious engineering effort. This was part of the Prometheus project (~30-35 years ago, not 20): [0] - https://en.wikipedia.org/wiki/Eureka_Prometheus_Project https://en.wikipedia.org/wiki/Eureka_Prometheus_Project [1] - The vehicle: https://driving.ca/mercedes-benz/auto-news/news/three-decades-ago-mercedes-benz-started-work-on-the-first-autonomous-car https://driving.ca/mercedes-benz/auto-news/news/three-decade...

erikpukinskis 8y ago
I'm just saying, there are teams meant to give it a shot, and there are teams meant to provide certainty a thing will get done.