Kids at home (going to home school math/science using scratch.mit.edu, and Lego Boost). The school sent us some great suggestions for a schedule, and activities so will make sure to balance out the screen time with outside time, and other types of learning.
Meanwhile, I launched the motorsports.ai website, will have the platform hosted there. Getting the beta ready, planning to use Hasura.io (GraphQL), and Vue for the backend/frontend.
The marketing stack will be Clearbrain/Amplitude/Segment, with BigQuery as the backend, and Datastudio for reporting visualization,
Continuing to make investments into improving my the platform, building out some customization for FogLamp to manage Driver/Suspension/Balance. Utilizing AI Platform Pipelines with my GKE/KubeFlow cluster to get everything organized.
On the more creative side, I am having fun working through some cool plug-ins in Ableton, and doing a Houdini/SideFX Foundations Tutorial.
Recently upgrade my DonkeyCar with a new diff, and a Raspberry Pi 4. Got it all set up with a cool retro bluetooth controller, really fun to drive, and train
To keep myself active, I have been biking, and running quite a bit, still able to get some private training in, and hope to get to a pool this coming week. Also, looking into a wetsuit for some open water swimming, looks like Austin Tri Cyclist has some for rental.
I’ve been enjoying the switch to eSports, and iRacing with both IMSA, and NASCAR this weekend, great coverage, and adaptability by everyone involved.
Looking forward to a hopefully relaxing week, spending time with my family, and working on pushing some of my projects forward. Here are some of the things I have on my list.
Road Surface Detection I have been working on a real-time AI/ML solution for predicting coefficient of friction based on the input from an optical camera. I am using an NVIDIA Jetson Xavier AGX (thank you ADLinkTech) , I have trained a SSD_mobilenet, and but would like to see if I can get some better performance (lower latency), and improved accuracy, so I am porting it over to the Google Cloud AutoML Vision Edge, this will allow me to test a bunch of different ML models automatically without having to run it through my ML pipeline based on Kubeflow, I am excited to see the results.
CAN bus CDDS, and FogLamp Part of the solution I have been developing uses CAN bus data to do validation, and verification. I’ve been using FogLamp to help manage the data streams, as part of this solution, I am going to continue work with Eclipse Cyclone DDS, and the ROS2 implementation, the plan is to get all of the sensor data from CAN bus to CDDS, then use FogLamp to send everything northbound.
Simulation Environments A lot of my testing happens in simulation, I started out with IPG which I am very happy with in terms of performance, and accuracy, however, I like to keep my options open, and have been working on porting my base model to the Simulink Vehicle Dynamics Blockset, while I was at it I decide to survey the landscape a bit more, and see what other platforms are available, it looks like some of the game engines are starting to provide robust simulation environments, so I am also testing Unity3d Simulation.
Spec MX5 Going to start working on the build sheet for my spec MX5, found a really nice donor down in San Antonio. I am going to drive her as a street car for a couple more weeks, hoping to get out to Harris Hill Raceway once or twice, to work on my heel-toe.
Family, work, racing, not necessarily always in that order, very hard to keep everything in balance. Over the last few weeks, I have prioritized getting ready for Road America, my race there is coming up over the Labor Day weekend, really looking forward to it. It’s been pretty tough to put it at the top of the list with all of the other responsibilities, and other activities I have going.
I had the opportunity to share some of my thoughts on this topic with Ross Bentley on his incredible speed secrets podcast this week.
Work has become increasingly demanding requiring more frequent travel, the partnership, sponsorship component of preparing for 2020 is a significant time investment, and the startup is gaining real traction with some exciting opportunities on the horizon. The near constant question I get is “when do you sleep?” Everything is a compromise.
I start by taking care of myself first, having a routine is the thing that helps me most, I guess I am a creature of habit, but breakfast, and training is how I start my day, when I am running, cycling, or swimming, I am able to organize my thoughts, and build a game plan for the day, it helps me get prepared, and regulates my mood, and anxiety level.
I have to focus on work through out the day, once I am home in the late afternoon, I make an effort to turn off technology, and spend time with my family, playing with the boys, making dinner, I love that we always sit down to have dinner together, and getting everybody ready for bed. Once the boys are in bed, I am able to have dedicated time for the startup, partnership, and work in the sim. On the weekends, time with my family is the priority, there are few opportunities during the afternoon, and evening to make progress on the startup, or work in on Road America.
Here is my method for working in iRacing, I am a big fan of data, so Virtual Racing School is a huge asset, I have a subscription that allows me to look at all of the data packs, and compare my times, with their excellent analytical tools. Additionally, I work with David Murry, and his webinars, he has a turn by turn breakdown of Road America that is incredibly detailed. I attempt to use the same approach for working in the sim, as I use when actually at the track, I take notes, look at video, and data, and then create a checklist of items to work on. This approach helps me feel prepared, and confident coming into the weekend. In addition to all of this preparation I have made it a goal for the season to work with a coach each weekend. I will be working with David on Friday, and Tom O’Gorman on Saturday.
A quick update on the partnership, and startup front, I was able to spend some time this week with Joe Speed from ADLink Technology, he has graciously offer the use of one of the NVIDIA Jetson Xavier DevKits, and Amit Goel from NVIDIA’s embedded team sent over a carrier, this is really going to accelerate the development of the models that we are running in the car.
As usual things are pretty busy, was able to take a small break, and get to the track this last Friday night. Edge Addicts runs an awesome event called COTA Under the Lights that starts at 3pm, and runs until 11pm. It got dark around 9pm, and both Danny, and I were able to get a session in the dark, turn 7 is epic. We made some good memories!
My old computer died, need one to get ready for Road America. So I finally decided to upgrade, got some pretty good kit, makes for a nice workstation, and looks rad with blinkenlights too.
I have always been a big fan of AMD, went with a Threadripper 2, and a newer NVIDIA card, I usually try to buy the previous generation, or mid-level spec, and give myself room to upgrade. This machine has 32-core, and 32GB of memory with a 2070RTX. I can expand up to 128GB of memory, and add another 2070 in the future if I decide to do some more development outside of the cloud.
We continue to work on proposals for partnerships, hope to start closing the gap by the end of this month, will hopefully be able to share more soon on that front. Training has been going really well, stepped up to 4-days per week about a month ago, and have already dropped another 10Lbs.
As is always the case, I have a lot going on, here is the latest. I was able to get the trackNotes app into enough shape to do a quick pitch and demo, which resulted in some early interest. I am ready to get some more feedback, and will be opening it up for a couple Early Adopters, if you are interested, drop me a note in the comments, and will send you the details.
One of my other projects is based heavily on computer vision, I decided to have some fun with it, and test my skills. I now have all the bits to build a Donkey Car it should be a pretty fun little project where I can learn, set myself up for the next steps of doing more suspension analysis work. I am now doing an extensive amount of model training using distributed inception approach, leveraging multiple GPU’s with a pre-emptible Kubernetes node pool. Good cost/performance, and translates to all of the cloud providers: AWS, Google, and IBM.
I have been working really hard with Phil, and continue to work my way toward my health goals. I have achieved 125Lbs of weight lost since October of 2017. I decided to add OPTIFAST back into my diet as a post workout drink, the high protein works to increase muscle growth, and helps with the recovery from a hard workout.
On the funding front, I am exploring a couple equity free options, there is currently a contest open with Google Cloud, and SAP called Circular Economy which is focused on United Nations sustainable production, and consumption goals. I am working out the broader implications of my current research, and plan to apply. Another equity free option is the SBIR route, which I have discussed briefly before. I have received feedback on my application, and have decided to pursue one of the current open solicitations with the DOT.