Building the FASTEST Self Driving RC Car

I built a self-driving RC Car that can follow racing lines and then tried to race it as fast as possible without crashing it. This time, it is WAY faster than anything you've seen in the previous video.
If you haven't seen the 1st video featuring this self-driving RC Car, you can check it out here: • I Made a Self Driving ...
Also, I'm looking for a Fall 2023 software internship, so if you're hiring, would love to get in touch. I built this entire self-driving car myself, imagine what I can do for your company 👀
Raceline Optimization Code: github.com/CL2...
Code ran on the RC Car: github.com/CL2...
My notes:
- stevengong.co/...
- stevengong.co/...
Credits
Thank you to Venkat’s group at Clemson University for providing the F1TENTH vehicle: sites.google.c...
My professor’s lab: yashpant.githu...
----------
MY SOCIALS
🌐 Website: stevengong.co
✍️ Blog: blog.stevengon...
📸 Instagram: / stevengongg
📱 Twitter: / stevengongg
☕ Buy me a coffee: ko-fi.com/stev...
----------
Subscribers at the time of uploading: 5,371
----------
ABOUT ME
I'm Steven, a university student currently studying Software Engineering at the University of Waterloo. I like to build cool things with code. I started this KZbin channel to share my love for learning and engineering with the world. I do everything ranging from electronics, CAD, control theory, programming, 3D printing, physics, and design.
My ultimate goal is to be able to make anything that I want. I'm working towards this by building increasingly difficult engineering projects that push the limits of my knowledge. Consider subscribing if you would like to follow my journey and learn alongside me as a student of the universe!
#stevengong #selfdrivingcar #fpvracing #rccar

[@stevengongg]

Thank you all for the support this past month, I really appreciate it! I got loads of exciting projects planned, so make sure to subscribe and turn on the notification bell so you don't miss out on my new videos :))

[@KonG-rk4od]

this is so cool. next up: add a ramp, see how far you can safely jump it inside and land it smoothly. math it out my man, you got it!

[@l_or_enzociotola8298]

So glad you introduced racing lines!!

[@ybaboy6403]

i am so glad that you are making evolution in racing robots !!! Love your memes in your video keep them in😂😂pls !!!

[@lakimakromedia]

I have seen, how AI in trackmania, adjust it own track line to make fast lap, maybe try to train, your own ai in trackmania, and then move it to rc ? :)

[@TheOfficialOriginalChad]

There are some "driving style" issues with your programming. I'm a driving instructor, so if you want to make it even faster let me know.

[@johng808]

i’d love to see this self driving car race against some hobbyists on an actual track

[@vancebente5186]

Exactly what I was thinking!!

[@xmysef4920]

That would be awesome lol

[@Hichilisthxforspyingonme]

This is a thing or at least was

[@TheFatAssCat]

The class of car might be a problem. This type of RC truck races on carpet or dirt which includes jumps. Onroad (racing without jumps) cars are very light and low. Could easily find a pro driver or hobbyist to race against this but I imagine you'd need a handicap as the equipment on this thing will make it very top heavy by comparison.

[@aalex1111]

@@TheFatAssCat I've seen videos of dirt RC buggy racing so I guess that would be the competition, but yeah this thing is outclassed as a platform already sadly

[@Pythnn]

If you really want to hate fine tuning, you could even start introducing slip angle, as on lower grip surfaces like polished concrete it can be faster to be putting power down earlier into corners and letting the tail of the car kick out a little bit. It would require gyros (which could also help with the oscillations at high speeds) but could also increase corner entry and exit speeds a fair bit

[@lukacrawford1619]

i was thinking the same thing with powerslides as in some cases it increases speeds quite a bit

[@lagule]

this is not just fine tuning this is transforming a mid-tier problem into an insane problem

[@NonJohns]

What I love most about these types of videos is that its pretty impressive already, but there's always a way to improve it, and more often than not, its a heavily explored topic

[@satibel]

@@lagule just use ML lol. jokes aside, that seems like an "easy" problem to solve with ml, simulate the car, then use the pre-optimized program control of the car with some securities and let correct for the delta between simulation and reality. (by easy I mean it's only a couple months of work)

[@xexzersy]

OH YEA SIDEWAYS

[@IngmarSolissa]

In the last few clips, the gear mesh sounds awfully tight. Also you can get better tires for it, so it does not slide as much. Final advice, the stock Traxxas servo sucks in keeping position imo, it's the first thing I advice people to change. I'm new to programming, so no idea how that part works (yet). But been racing and working on rc cars for the last 15 years. Love these video's, gets me excited

[@Mow3829]

I agree with this, i was hoping it would already be in the comments!

[@a5th3n09]

Yea that gear mesh was sounding spicy

[@3rdDegree_]

Yea you definitely don’t need that much tread on such a smooth surface, braking and acceleration would be much better

[@JoeOvercoat]

Get an INJORA servo for a great price point performer. That aside, I suggest the mesh was not getting tighter but rather we’re hearing some tooth wear. He is bashing, after all.

[@STRIKEcorperation]

If Steven ripped out the guts of the servo and replaced it with something that could output the steering torque and actual position, it would be possible to feed the controller a "force feedback" that could be used to "feel" the track as well as allow the servo control loop to be tuned dynamically through software.

[@CrueMusic]

Absolutely epic! This motivates me to start a similar project

[@stevengongg]

Go for it!

[@Talaxianer]

The $1600 LIDAR sensor motivates me not starting it 😄

[@StuckOnAFireHydrant]

​@@Talaxianer could always go the musk method of 2 cameras for depth perception and pray your model doesn't floor it into a overturned box truck

[@dave.h]

@@StuckOnAFireHydrant everyone knows self driving needs LiDAR except Musk 🤣🤣🤣

[@Anil-bg5se]

​@@Talaxianer use a lidar from robot vacuums. They are 150.

[@npsit1]

To be fair to you, people have other senses to rely on when driving. You just have the LIDAR. People can feel when a car is losing grip or how hard they're accelerating or braking or how close they are to something. You've done a really great job with limited sensors on your vehicle.

[@hyper_channel]

Exactly, this problem is nowhere near as trivial as it might look, literally years of research have gone into driving simulators only to get the grip formulas right, apparently there are pretty good models for dry and high grip situations but when traction is lost is not yet solved (assetto corsa devs have videos talking about this in depth). A pilot is getting A TON of information from G-Forces applied to their body, so that would be a great sensor to add to the car and formulas.

[@Habberplabb]

Dope project! I'll be on the lookout for follow ups. An unwarranted thought: You can probably squeeze out more performance from your controller with something like a model predictive controller with constraints for following the racing line, vehicle dynamics and local obstacle avoidance in your lidar map

[@Christian-mf4jt]

You might not even need to run the MPC in real time: Just use the existing optimizer with starting conditions close to the racing line, and linearize the necessary controller output around the optimum. Calculate and store the parameters for every point on the path, then use them to get an optimal linear controller at every point along the track.

[@VulpeculaJoy]

@@Christian-mf4jt With slight deviations accumulating, eventually it would get out of sync with the pre computed inputs. There are hybrid approaches, where one MPC generates an upper velocity bound offline and a second MPC tries to reach that limit online.

[@JoeOvercoat]

@@VulpeculaJoy The missile knows where it is, thusly… kzbin.info/www/bejne/mWqXk3SCdsx4hK8 …so yeah, a map if one can generate one.

[@billybagins8391]

Bro u basically copied this entire thing from the MIT 6.141 class on robotics (now titled 6.4200). Please mention that.

[@Devthedev.]

true

[@TaiViinikka]

Steven, I'm wondering if the LIDAR is doing anything during your speed runs, or if it's just used to help build the map? It seems to me you are doing autonomous driving via a sort of multi-stage process where the vehicle doesn't have any capacity to respond to obstacles, changing track conditions, changing internal mechanical (like gears eroding or tire wear.) Don't get me wrong, it *STILL* looks damned hard and we're enjoying the details you're sharing! I think it might be helpful to spell out for the viewers that this is (self)-mapped autonomous rather than real-time autonomous driving. Also, if that LIDAR is a snap-on module and isn't doing anything during racing, maybe pull it off and save your $1500. :) Please let me know if I've misinterpreted your process. It looks super fun either way. Glad you had some friends to help out with the 2am laps.

[@stevengongg]

The car is actually using the LiDAR sensor to localize itself as its racing, which is why it's needed. The code for that is from this repository: github.com/f1tenth/particle_filter You are right that it doesn't have any obstacle avoidance or can adapt to changing track conditions, so this is definitely something that I want to work on in the near future :)) one step at a time is my focus

[@luisr320]

If the car was only following a preset line and not the lidar data to define it's position along the circuit, any slipping would offset it from the planned line and very quickly it would be on a collision course with an obstacle.

[@joshseal2000]

@@stevengongg this would be great to see. Specifically, how narrow of a gap that hasn't previously been mapped can you get the car to drive through at speed! i.e. you add a few barriers around the course that the car needs to drive between in order to complete the loop. You are totally right though - start simple first. Good luck

[@csehszlovakze]

@@stevengongg at least implement some emergency braking if it's about to go into something or if it's getting out of control. later on you can turn that into collision avoidance based on previous knowledge of the area.

[@richardwackeroff4897]

@@stevengongg REALLY COOL PROJECT! I'm not a racer, engineer, or coder, but I thought a cool idea, using what you presented, would be comparing optimal drive lines on existing tracks. you could take an existing track with Ariel photos, sometimes they have track data, and compare your code for the optimal drive line vs center line vs recommended lines from the different race teams. you could hop on that AI learning train and see if an AI can spit out anything that beats your program or existing lines for the track. if you really wanted, you could go down the rabbit hole and start scaling the rc car to comparable race cars that have driven on the track and you can build the track at scale (somewhere idk? space would be an issue)

[@wanderingcameraguy]

As an engineer. I absolutely love this and I am inspired to make my own.

[@5MadMovieMakers]

Gotta go fast

[@frfrongong9712]

You should totally make a drone that follows it around like a third person view racing game

[@jatre5938]

Stoopid

[@JoeOvercoat]

RF tag babeeeee!

[@DaveSmith-cp5kj]

You don't need a drone, just a 360 camera.

[@ironman5034]

i noticed in your optimization parameters you didn't include the car mass/inertia, that can help reduce the unplanned overshoots (if your PID is optimal anyways)

[@hyper_channel]

yeah it feels like there is a bunch of low hanging fruit that could improve the overall performance quite a bit, I understand they might not want to sink insane amounts of time and money on it though

[@Alien_Jenkins]

Maybe it's worth making a "safety square" around the car so that it avoids obstacles in priority over the trajectory of the track

[@gkelly]

UWaterloo! Those new buildings have a lot more windows than the previous uh... 'aesthetic'. This is such a cool project. I remember trying to write the firmware for a fully autonomous drag racer one year in-between exams. It didn't go well! Your results are awesome!

[@fynnjackson8416]

For those floors I’d recommend slicks they have a higher contact patch than gridded tyres although slicks aren’t great for anything other than smooth surfaces. Also reducing the sensitivity or speed of steering to reduce instability

[@gstrnerd2]

Also, wiping tires down with simple green adds some tack to the rubber.

[@renatobfa]

Congrats!!! First thing, in my opinion, would have been to change the stock tires. There are some that are so much grippier then those, and would make the car feel much more precise.

[@blacklangster1141]

Also! A great addition would be slick tires for more grip on this smooth floor

[@jessedennerlein9746]

Wow, that is so cool. I built something similar 10 years ago after I took a class from Sabastian Thrun from the Google Self driving car project. I couldn't afford lidar back then, the cheapest I could find was 30,000 dollars so I'm excited to see that there is now a semi-affordable lidar unit. Way to go. Keep going! Here is the video of the one I built back then. Yes, I'm using an Iphone to get a compass heading and a GPS location signal. kzbin.info/www/bejne/qYHaqox8bsl6jpI

[@DSdvdDS]

This is probably one of the coolest projects I've seen showcased on youtube. Editing for the video is fantastic too! Keep it coming!

[@gracehopper9549]

I'm making a racing game, your notes are a LIFESAVER my man. Love the videos, I'll be following and not just to scoop more insight.

[@Videolinquency]

If the government had been able to understand how fascinating this is, they would no doubt have taxed it. Thanks a lot for the ride!

[@Mazzphysics]

So hilarious watching robot vids lmao. I remember building a mapping robot when i studied, we only had ultrasonic sensors, which were garbage in comparison to lidar. But still nice to see your robot crash randomly :), just like ours did

[@Flimzes]

Ultrasonic sensors might have mapped the glass sections more effectively

[@redles4212]

I don't know if you are aware lf this or not, but that particular rc car on a 3s lipo can go about 60 mph with the proper gearing. So in the future you could potentially go very fast.

[@mateoichijo3408]

Man this is such a great content! Im currently doing the same project with a go-kart using Stereo Visual SLAM. Good luck bro!

[@stevengongg]

omggg lucky! i wanna do the same with a gokart, but those things are EXPENSIVE

[@TheSwedishGamerTim]

This is definitely better than tesla

[@L1EcmJkFwm8]

I think you need an accelerometer to determine the G forces in corners, so you can adjust the steering and acceleration to prevent under/oversteer. Nice results by the way!

[@gavinhogberg]

Michael Reeves inspiration and Initial D references. 10/10, subscribed

[@-Rober]

Your videos are amazing! Nice to know how you manage to get over all the problems you found and all the solutions you made for this. Do you know the existance of the AWS DeepRacer League? It's more like image-processing more than LiDAR (also used for the obstacle racing) but you can find hours of fun out there

[@DizzyDJW]

You didn't need to wet the floors you could have wet the tires instead, called "Tire Slicking", it's illegal in NASCAR but who tf watches NASCAR anymore? For RC Cars you can usually find "Traction Compound" at your local hobby shop that sells RC Cars

[@Pootie_Tang]

Man, you have a great channel, I like to witness interesting and non-trivial ideas being brought to life! Keep it up like this and pretty soon you'll gonna be big!

[@dinoscheidt]

The tires are just awful for that (but they clean the dust from the floor). As a machine learning engineer and a rc enthusiast (have the losi dbxl-e 1:5 scale), not even a good human rc driver could race that. You’re just skitting at random intervals due to dust accumulation. So suggestion: 1. Buy some RC motorcycle tires to reduce the surface. Maybe even lego motorcycle tires fit. (Alternative: Pour Asphalt on the 5 floor 💪) 2. Maybe (momentarily) lock the drive wheel differentials (or don’t forget to incorporate them into the simulation in step 3.). The problem is, that you fly out of corners because one tire eventually loses grip from the flat floor, creating lateral forces throwing you around. That helps you when you have fast reactions times as an experienced driver, but compared to locked differentials removes predictability. When you lock the differentials, you’ll see how hard to drive it actually is on that surface. 3. To get the ground truth into the simulation, it probably makes sense to throw the track into unity3d, grab a standard car rig model and have a random tree search or even ml agents tweak the coefficients of that car rig so that the simulation matches what happens in the real world. From there you have a lot of options. Like ML Agents / Curriculum Learning etc. best with stochastic deviations over the coefficients so the model learns to deal with none-perfect environments. Oddly enough, you don’t have a perception problem but a control problem. Usually its vice versa for autonomous vehicles. Good luck!

[@UnitSe7en]

Don't mash the gear teeth together as tight as they can go. You need a minimal amount of backlash. After your fix it sounds like the mesh is too tight. On the chassis you have, (TRAXXIS or clone) it's a very fine adjustment to get the best mesh. Sometimes slipping some paper between the gears helps. You're drawing to much current through the ESC if the meshing is too tight, because the motor finds it hard to spin. This can damage your components.

[@windsorpiano]

"if you can play something slowly, you can play it quickly" generally, is an outright lie. It's much more efficient to play something faster than it's supposed to be, then slow it down to its proper speed. Also, learning music is less boring this way.

[@codaman127]

I'm so stoked to see this work out, but boyo, please set a ceiling on your audio mixing. You just Mario Kart N64'd my eardrums to death.

[@KevinGre]

When the car is off the ideal line, do you just control to get back to the optimal planned line? That is not what a human would do. Human drivers will, on the fly, create a new optimal line based on where they find themselves. You drive the line you find yourself on, not dive to get back to the optimal line (which can overshoot as you saw in your oscillating). The solver probably takes a while to produce the line, so I'm not sure how you would get that into the control loop. Without adapting the plan for the found grip levels (do you have more grip than the plan assumed or less?) and for errors (like understeering in the corner entry due to less grip than the plan assumed) you can't go as fast as possible and will always have to be conservative on the grip level the plan assumes. The great drivers probe for grip, often in braking or even corner exit where overstepping the grip won't cause a crash, just cost time, and then use that information to adapt their plan. Competing with intuition is tricky. There are also different techniques that the control loop can use to keep on the planned line, like using power changes to induce weight transfer and hence changing the grip level on each wheel. Such as what drivers call throttle steer. It would be interesting to see how much time is gained by being adaptive to the conditions vs adding better car control. Thanks for sharing your experience.

[@SkylordBrebro]

I find this really cool, BUT, I want to see this race against a human. Attaching an FPV cam and get goggles to wear for some "professional" racer to try to see how they would compare against the computer.

[@BarakaGameDev]

Hey, make a 3rd person follow camera drone. Like in the video games

[@chemicalvamp]

Dude that is awesome, Can you implement drifting? If this is what your class does, it needs a youtube channel :D

[@nicolasbertozzo2997]

What if you let us control your car via internet?

[@rctestflight]

This is really awesome!

[@stevengongg]

OMG I LOVE your videos man :D

[@bigbaderek1978]

why dont you 3d print a tpu bumper for the lidar? i can do it for you if you don't have access to it, I just need measurements

[@anshtanwar1813]

You must be knowing about aws deepracer league where we have to solve similar challenges. Can you make a video related to it

[@lukas823]

This gives me Michael reeves vibes, but different and I'm all down for it

[@Zelk1]

This thing is supper cool now add a jump and see how it will (crash) i meant "work"

[@qwertyuiop-ux6jk]

fantastic work! it reminds me fastest maze solver!

[@makermandan]

Really nice video! You earned a sub :)

[@TheGmodUser]

Why are you slamming all the time? And not switching to just localisation?

[@captain_rex896]

2:16 I liked the Bionicle reference, when you showed Mata Nui )

[@IOUaUsername]

Constraining yourself to tyre grip is not the only approach. By using gyro data and comparing it to steering angle (hint: tap into the servo's potentiometer signal so you know the true current steering angle rather than the requested angle), you can determine the oversteer or understeer angle. Oversteer means the car should countersteer and slow down. Understeer means the car should just slow down. Using the oversteer ar understeer angle to augment the determined direction of travel from the camera/lidar data would allow you to eliminate the oscillation that wheel slip introduces to the control system. You could alternatively use an optical flow sensor from a computer mouse under each axle to for more direct and lower latency data which is less susceptible to noise than an IMU.

[@mw2filmmaker]

It sounded like the gear was slipping again in the last run. Needs to go fast fast fast!

[@jimcurt99]

Never thought I'd see a Slash do that! pretty cool

[@ambushell5778]

great work! with the right tires, less weight and a lower CG chassis you could reduce some of the understeer the car has and increase the speed much further. I'd suggest a touring car chassis like the Tamiya TT02 or similar

[@dannyb9083]

when an engineering student actually does their homework

[@torbjrndyring9971]

now just learn it to drift/powerslide and that thing is fast AF

[@OniKidou]

As someone who works in the field of autonomous vehicles, I have to say you did a banger of a job with the localization/mapping, and programming the annotations for the RC car to be able to race around the hallways like a pro. I can't wait to see it do more though, like detecting and avoiding obstacles.

[@slendi9623]

Now add drifting abilities and it will be truly amazing!

[@guysawyer3844]

This was so cool to watch! Took me back to my late night mechatronics days, in the lab till 6AM trying to figure out why my motors where going the wrong way and replacing H-Bridges haha!

[@BimmerWon]

I thought it was the Adventure Line in the thumbnail.

[@civicrocker]

One thing I’d try is better tires with a stickier compound. That would help a lot with the spinning out on 100% launch.

[@StephenYang]

Softer compound slicks would be ideal.

[@NicholasRehm]

Totally my kind of project!! Awesome stuff man

[@InYourDreams-Andia]

Very cool! I often thought about this approach, but from a different angle. Instead of fully autonomous, I think a better approach would be to teach the car a track. The receiver and transmitter record a track in slow mode (driver inputs). Then parameters are set to for full speed and corner speed. You'd need a gyro to keep it in a straight line, and guide the car around, each lap finessing the course speed etc. So in racing mode, (hands on still) it would auto correct, hold direction and find the best line for each section.

[@gstrnerd2]

Gyros are great at measuring inertial change. However, it would be pretty difficult for them to be used for localization alone.

[@InYourDreams-Andia]

@@gstrnerd2 true, but if it's set to counter steering anomalies, it'll stay on the programmed track

[@狮子王-q9k]

好视坚定支持Steven

[@hoopengo2289]

2:15 - Я секунды 3 непонимал откуда в видео русская речь ахах

[@AerialWaviator]

Nice work, keep pushing the limits. Both the car, and yours. Was surprise that your racing on the track counter-clockwise (12:32). Can also test racing clockwise, it would be like having a second track. Would be a good test for your algorithms, once you think you have the car tuned for one track. (might also good for evening out the tread and steering wear ;)

[@tristanbased-af2159]

hey look ricegum is smart now lol

[@UrBoyDlo]

Bro this is so dope you are making me pursue my dream of coding

[@elliotmarks06]

Now you just have to teach it how to drift

[@depressedbread6893]

Now make it drift around the corner

[@Cryptic808]

2 Thoughts, Steven: 1. Looks like you did constrained training to find optimal racing line? I wonder if there is a simpler way based on physically simulating a bendable wire like a bendy coat hanger material. Or using Procedural splines. Meaning. If you shrunk your track down and made it a 3D model and you needed to put metal wire inside it, that wire would automatically find it's optimal angles through the whole thing (I would think). 2: I know others have brought up better tires and parts, but I think you are leaving a lot of speed on the table with your current setup because of your understeer issues. In racing , Rally drivers use a method called the "Skandanavian Flick". It's a way of steering away from the turn in slightly to shift the weight to the back to rebalance the car after hard braking has shifted all the weight forward. Your car is "pushing" into the walls because it's letting all the weight slam forward and it doesn't clean it up with a pre-flick. All drivers use this flick, but the more grip the car has the more subtle it is. But if you can code in the flick it should balance your car where it can make turns better and if it slips it's a balanced "drift" and not a push into the wall.

[@bummers]

This is so awesome! Reminds me of my undergrad days doing control and instrumentation. btw, wonder if your motor and steering control code take into account the grip and slippage of the tires as this will determine the effective acceleration and deceleration the RC car is capable of. At least that's what actually racing teams do.

[@fewminutestudio959]

I think that's a little out of my mans skill set.

[@jorgeestefania3184]

Now that ir knows how to race, let’s teach the car how to drift

[@banovsky]

Car missing the apex by like 3 feet on each turn lol BUT SERIOUSLY, amazing work, this project is fantastic. If you have karting in your area, might be a good place to both practice your skills and give the RC car more run space on a better surface (when it’s ready ;) )

[@eckee]

How much did the car cost?

[@enginediy]

Cool

[@shashimalsenarath4146]

nice work man

[@isntitawesome2104]

this is not true self driving.

[@chillieuwu]

You don't even know what self driving is then

[@isntitawesome2104]

@@chillieuwu when a car can go in any road with the help of it's vision then it's self driving. If you have to drage a car with a lish in every step it's not a self driving car.I think in this day and age everybody know what is self driving and how it work and this is not it.I found this video when i was working on my own selfdriving car with traffic light detection,lane detection.i find this video to be lickbate. A junior developer at my office can make this sh**.

[@justinhealey-htcohio3798]

Hi, this really is some truly exceptional work. I wish I learned to program but now I'm almost 40 🙄. But, on another note, I've been obsessively following the Ukraine war everyday since Russia invaded because I have a friend there.... So, when I see this kind of project it makes me wonder why you & others don't used some of your brainpower to help Ukraine make the thousands of FPV Kamikaze drones that are holding back the Russian Occupiers on the front line?Specifically: 🔹Overcoming GPS & electronic warfare jamming 🔹 Creating an inexpensive solution to help triangulate Russian Artillery through acoustic direction finding. 🔹Passive Radars 🔹SLAM Lidar to autonomously locate Russian bunkers 🔹A solution to identify and take out Russian cruise missiles and Big Kamikaze drones.. So many possibilities are in your brain! 🔹 Russians are hiding with machine guns inside holes in the ground along a 2000km Front line.... Just imagine how much good you could do for them!

[@IbericaOverland]

Awesome work, but to get better results you should work on the mechanic part too, those tires are for off-road, and your suspensions are too soft 😊

[@matteob5200]

Can you make it drivt

[@anubis520]

Wow, excellent improvements over the last video. Perhaps start in a slower part of the track to prevent the car instantly losing traction and hitting the elevator. Great job with the project. I look forward to the next video.

[@IonNight]

The pinion and spur gear mesh sounds bad in the end. When you're setting the gear mesh you can put a strip of paper in between the gears before tightening the motor to get good mesh. You might want to add a heat sink with a fan to the motor and the esc. The Traxxas VXL-3s esc should work fine on NiMH and 2s Lipo without a fan, but if you're running 3s Lipo, you should use a fan, but to proof the car, I would add a fan to the esc regardless. Like someone else said. Get wheels for street. Traxxas Anaconda are okay tires, but I have never tried them on linoleum floors like it looks like you have. I have the same car, but the LCG version, Im running a 2650kv motor on 4s. Top speed is about 120 km/h.

[@xnadave]

Outstanding work! I'm wondering how well this would work with an AI model. Inputs would be the lidar image, current velocity and steering angle. Output would be steering and velocity adjustment. (I predict this would fail, but it'd be a spectacularly awesome failure. )

[@casadioDesign]

Hi Steven! I just saw this amazing project, awesome! I sent you an email to share with you my full 3D printed rc car, it will be awesome to make a homemade 3D printed full autonomous RC car!!

[@MateuszRulski]

I love the project! Wow! But "self-driving ... remote-controlled" sounds a bit like an oxymoron, doesn't it?

[@ΓεώργιοςΠαπαδόπουλος-φ4υ]

Great project and video! What is the top speed you achieved in the big track?

[@theabbie3249]

now we need it to drift at sharp turns

[@Soken50]

This brings memories from mecatronics engineering school, well it was more spidery eldritch horrors trying to learn how to walk using ungodly transform matrixes but it was just as ridiculously fun at 3 AM in the engineering department. We could even swing by the agro-industry department and taste test their latest "original" creations.

[@Cutie22704]

Teach it to drift!

[@JoeOvercoat]

You set the goal much too high. “F1 driver” is beyond attainable. I suggest the vast majority of people simply don’t understand the accomplishment level of an F1 driver. P.S. That gokart operator should be leaning out on the turns because he’s driving a fixed differential vehicle, where one needs to take weight off the inside wheel to allow it to slip instead of drive the cart to the outside of the curve.

[@urnoob5528]

when will it drift?

[@luisr320]

The next stage would have to be the integration of an IMU to stabilize the tracking. You should think of what an actual race driver must feel in order to optimize it's performance. If you seat a driver in front of a computer screen and strip him from physical sensations, the car dynamics will be lost and the performance will degrade accordingly.

[@SkyFpv]

Do yourselves a favor and pick up a cheap analog FPV setup. That way you won't have to chase it around. Also could be a nice foray into vision based driving ;)

[@RayRayWasAGoodBoy]

I think you could squeeze out more if you added a body for downforce and better tires for that surface. Cool project!

[@VanWarren]

breaking the LIDAR simulates a racing death, which is the next logical step.

[@GetRealwithMike]

I used to race Formula Mazdas and I see why you are having problems in your corners.. Let me explain. As I approach a corner at high speed, I use two feet - one for the brake and one for the throttle. I hit the brake as I approach. This shifts momentum and weight to the front wheels to provide better handling and dig into the corner. This is really brief. Then I go into the corner accelerating through the corner. This prevents the car from rolling out (to the outside of the curve) and makes it stay in the groove in a forward direction. All weight and power are pushing through the corner instead of fighting to keep the car from flying off the track. You're trying to control the flying part instead of making it push through.

[@cocobos]

Next, crowd driving AI, would be nice to see a flock of AI racing cars in sync and overtaking each other. Could be a thing in the future for kids.

[@shirubaurufu]

Is the Lidar being used during the driving? If it's only being used to map out the track to generate the racing lines, you could reduce weight (and risk) by removing the lidar when it's not in use.

[@spencermiske250]

I got into UW mechatronics a bit ago, but im scared of how cutthroat the program is. Seeing you have fun with this has made me want to go!

[@jonathanr4242]

That is awesome. I tried hacking an rc car for autonomous driving and my friend called it “the skirting board destroyer”.😅