Author: rb0unis

I had no idea if we would be able to get it into the lab or not so it sat in the foyer for a while I ordered some jacks, axle stands, and skids. Eventually we decided to move it into the lab. It took 3 attempts to get it around the corner in the corridor we were just about to give up when someone suggested one more attempt. Not sure how we managed it, but we did get it turned so we could push it down the corridor. We filmed this as I figured at some point Id probably end up making a video.

The car was taken off its skids and reversed into the robot lab.

Then not a lot happened for quite a while, largely due to teaching and other research projects. However, the px2 didn’t arrive until July ‘18 and the LIDAR didn’t arrive until in Jan 19. But a lot of thought and research went on. Action was a result of the pending CVSSP 30th year celebration and being reminded that the car was coming up to its annual service (and Id probably only put 5 miles on the clock). In discussion, Oscar said that if we wanted to get it done, then we were probably going to have to come in at the weekend. Any spare time (and then some) was being taken up with the new robotics module. That weekend I came in and spent my first Sat morning cutting a roof rack to size…

Very early on I decided I wanted to document the build, maybe create a website or blog that provides details of the build. But as grants started to get written I found myself making promises to make the entire build open, with instructions for both our hardware setup and the software that we develop. I’m not bound to any of these promises it seems like a good thing to commit to. So as we started work on the car I kept saying Id start a blog. Its now 12 months since the car arrived and Im only just starting it but that’s because we are only just starting in anger with the bulk of the work and although I have the best intentions I apologies now if this blog is sporadic or tails off, only time will tell….

So back to the car….

Its Feb/March 2018, the car is on order and Im starting to spend a lot of time thinking, planning, scouring the web for info. I decided we should go with an nvidia PX2 drive with the suggested Sekonix cameras. Not sure this was the best idea as we will see later but at the time, it seemed sensible to go with the cameras that nvidia suggested as I knew this would avoid compatibility issues. NDAs etc were put in place with nvidia and that was ordered. Then it was time for a LIDAR. We are primarily a vision group but LIDAR was important 1) for safety reasons and 2) because its a key component of most autonomous vehicles. An obvious big name is Velodyne but they are not cheap. However, there was a press release from velodyne that they were dropping the price by half. So I contacted them and they put me in touch with the UK reseller. I asked about the 50% discount and was told that was only on orders of xxx thousand, so I asked about an educational discount and got a blank. To be honest I wasn’t that impressed, it felt like unless I was after a bulk purchase they were really interested and that made me look for an alternative, as a matter of principle. I found Ouster, they had some great discounts for education and were therefore much cheaper than a similar spec device from other suppliers. Ouster it was, and this also ordered. I have to say, while the Ouster is great, feels like we got a lot of LIDAR for the money, it took a long, long time to come as we had to wait for a new production batch. The LIDAR finally shipped xmas week and it was waiting for me Jan 19.

The px2 arrived summer 2018, we were all very excited. Actually that’s a lie, Oscar and I were very very excited.