n @AecertRobotics i wouldnt use CA glue as it leaves smoke like effect on the parts thats not easy to clean. i use UHU all plast glue. it is great for plastic especially the printed parts
I have a few ideas Add LiDAR and make it map its own environment Add a nerf gun attachment and a button on the controller to shoot Program it to be able to walk up and down stairs
Give it the body of a turret from the Portal game series. (Also loving the design so far, I know nothing of this sort of hobby, but this is quite enjoyable.
A cool yet challenging attachment wuold be putting some electrode on the tip of the front legs, the one it uses for the slam attack, and wiring them with a high voltage booster so when it attack it can "zap" the target. you could also put some metal spikes on the already existing leg shields and do the same thing
Some (very far fetched, but oh well) ideas: so if you’ve ever watched avatar 2, they have these swarm robots that build things and they kinda look a bit like this however they can walk up metal structures, could you add a small electromagnet on each foot? You can then power them when in contact with metal. Another idea for holding things could be to add another servo to the central pair of legs to flip them 180 degrees so they face up. Then they could have slightly wider feet or grippers so they can pinch whatever they are grabbing together. Kind of like a human holding something over its head. Some sensors to detect objects around would be cool especially partnered with 3D scan software to map or upload an environment. Finally, the easiest idea, put a pen on the bottom and map its movement to draw something
I was searching some project that have a spider shape (Hexapod) to materialize an idea I had but not even in my best dreams I could imagine find something as awesome. Congratulations. I'm going to github to get all details about your project, thanks for that. If you have some tutorial or if you decide sell an kit about this project as you comment before please let me know. For a while I'll printing the parts shared by you and study all details about the assembly. My idea to this device is include a camera 180º to do some work at height on metalic structures using magnetic parts in the legs. I don't know exactly how do this but is something like that I have in my mind.
you can easily screw directly into 3D printed part, just use self tapping screws, I've used this method on my 3D printed robot dog and it worked flawlessly
You can attach some devices on top like maybe a flying drone to get a different view of something and then when the hexapod sees or senses something and needs a different view to accurately understand it then it could. Maybe put a portable charger on it and a phone pad and then lay your phone on it and then youll never lose it because you can call it and bring your phone. Attach a speaker and mic that are small so the bot could understand speech and talk back like siri or connect an ai. A cam with a connection to ai would be cool too and help identify things and assist in work and finding things.
Great video and project overall!! You probably could make the snap fit work if you rotate the orientation of the way you print the part. You can easily separate and snap across the layers of an FDM print, but if you change the orientation of the print so the bend direction goes along the "length of the filament" it gets a lot stronger.
I actually did try this, but the shape of the peg is extremely important, so it had to be printed right side up. Printing it sideways made it so the peg didnt fit properly into the hole. 100% right though it would stop the clips from snapping. I think im going to go with glue for this sort of thing from now on.
@@AecertRobotics I also was going to say this (and I did pause the video so maybe its something you haven't addressed yet). If the problem with reorienting it is that the corners over bulge, thus ruining the fit you could try a small nozzle or only having one side be the "flexing" side, thus giving it a larger size to work with. But im assuming there is a reason the snaps are at the size you showed. Honestly the screw holes in the armor look sick so besides making it a bit cheaper to reproduce the snaps don't seem that important. Mind if I ask where you get your heated inserts and screws from? I see plenty of options on amazon, but some have way too many variants to make it worth getting
I build a teaching tech hexapod ages ago. It worked ok but this one is faster stronger! Well done I'm learning a lot from your vids. Btw an arm that can grab things would be a great attachement. For something simpler maybe some distance sensors for autonomous walking. 😀
Depending on where you go with this project, a food dispenser would be hilarious! if I remember right, this robot is controlled by a person, perhaps a camera attachment could help you traverse easier. I'm thinking of a camera that can rotate 360, or by some margin, as a camera locked in place wouldn't be very flexible. I'm just blabbering really, awesome work on the robot, those attachments are looking mighty fine!
The form function in Fusion 360 will help you achieve those organic shapes like the armor plates much more easily. Also you'll find that for someone with a background in polygon modeling it's quite intuitive to work with.
cool attachment idea, tho this might be as big a project as the hexapod itself, but with fusion 360 at your side, a lot of it may be easier than it sounds at first: design a simple mechanism for firing airsoft pellets in full auto, could be as simple as a hopper feeding pellets thru a tube that becomes the barrel, with small gaps for spinning wheels to send the pellets flying, like hotwheels toys of yore, so that the barrel can be really short, allowing you to pack it into a housing that follows the existing portal design aesthetic of the leg armor, so it's like a walking portal turret, you could prolly control it's firing angle with a joystick and fire with a momentary switch bonus points if you slap a camera on and make it fire at anything that moves, saying lines from the game thru a little speaker, like asking if anyone is there after it stops shooting, cause that would be hilarious🤣
I kinda had a bad feeling that you got bored during the process and gave up on the hexapod when it was unfinished. Nice to see you’re back in action Edit: as for the ideas for attechments, i don’t know how much of a good idea this would be, since it won’t be functional, but maybe you can build an oval shaped head that carries replica infrared scanners and railgun and gatling gun mounts. The head can rotate x axis and the weapon mounts can rotate y axis, like a gimbal. This can make the hexapod look like a boss fight foe, that is out of a game with a cliche story where the hero fights with an evil genious, and has to battle the hexapod in his hideout to reach him.
have a robot arm on the hexapod. With a robot arm (while more weight) you can instantely add a lot more functionality to the robot and shouldn't be so hard if you had already made a hexapod. With the robot arm you can make it grab anything or do anything with the right attachments to the arm.
I think putting propellers on the back of it would give it of a challenge and make the movement 3D. When you get the landings down you can try and use it as a flying hook.
Add a big blow up hammer on top to hit stuff with and put a bit of amour on top so we can have battles with a sesneor which counts the amount of times hit
looks amezing, especially with the armor plates. my idea is stupid, over complicated and probably won't work... but, I would love to see a (single shot) nerf dart turret on top of this hexapod hahaha
Heated inserts and screws isn't the only way to make the armour rock solid, a decent epoxy (or hell even a decent CA glue) would probably be stronger than, or close to as strong as, the PLA
I should preface this as such: I don't usually use PLA+, I usually print with normal PLA or sometimes nylon, as well as a lot of resin. That being said, for CA glues even the cheapest stuff my local dollar store has works well (I've glued everything from ABS, PLA, resin, wood, my desk, and my fingers) so I don't have a specific brand recommendation per se -- if you live in NA (I don't anymore) gorilla glue should be easily available and fairly good, loctite super glue comes to mind as well and should be available almost anywhere. As for 2-part epoxies, again even the cheap stuff usually works pretty well. But loctite springs to mind yet again as making pretty good epoxy glue. 3M makes a good CA glue as well, come to think of it. Pretty much any brand known for making glues, for more specific recommendations I'd probably suggest you google around a bit for what works on your specific brand of PLA+ (formulations matter, and I don't use them on a day-to-day basis so I can't say for sure if I've ever glued any PLA+). Or you could just print some test pieces and try different glues. It's worth noting that some CA glues can take hours to fully harden, especially cheaper ones. I usually hold the part in place for about 10-20 seconds until the glue sticks it down good, then leave it overnight (about 24 hours or so, let's say) and it'll be rock solid. Epoxies vary, but usually have longer open times and set times and I'd consult the instructions supplied with the specific glue.
I've been looking to find a good set of components to achieve this and at the price you paid, the BEC is actually a good choice. It would be challenging to achieve the same robustness at that price point. Also, it's the most likely part to fail in the circuit so having it external is not a bad idea. I have used the hextronic YEP controllers before and they are quite good.
@AecertRobotics this project is amazing. i was following it from the start. I ended up building it. the build went smoothly, got all the parts and electronic, connected everything. there was no smoke luckily. unfortunately, i was unable to get pass the calibration state. one servo ended up dying. fried completely from inside. I only need help in this stage if anyone can help please. just P.S. i didnt add the current sensor. there was nothing in the code that suggests that it was in use. I just bridged the too current pads to override it.
I really love this idea. It's out of the scope of this project since it would require way more processing power, but if and when I ever make another hexapod with more advanced electronics, I will 100% be doing this.
one idea i'd love to see mounted on the hexapod is a cannon with a cammera on it that would allow you to control the robot, aim to a target through the camera and shoot the target, let's say a can of soda. it would be awesome !
or it would be cool if it could have a top mount for a mini projector so that it can project pattern on the ceiling or even just a ring of lights that can express things like when the battery is low, when it's starting an attack or charging (if you are gonig to make him rechargable without disassembling). Obviously the projector could make all this things other. Sorry the spam but right now i can't realize them on my own so seeing them wuold be really cool
If you want more inspiration look at Mark Setrakian history of robots, specially Mechadon, a 500lbs battle bot he rode in on. Though this battle bot wasn’t ideal for battles, it was cool. For cooler ones look to The Animatronic Robot Designs of Mark Setrakian in Adam Savage’s Tested. See the hand gently, though fluidly spinning a globe. I think your design is really cool and also reminiscent of his designs
love the vid. I remember you mentioned a full course including coding in the first video, really looking forward to it. Care to share some details about the course?
Probably some set of sensors for surroundings mapping and trying to learn hex how to navigate in space and time. It‘s pretty interesting coding challenge as well
Another UA-camr, the 3dprintingprofessor created a good snap fit connector for his printablok project. It would be cool to see it used on your hexapod.
Love this series, been super interesting to watch. I don't know if you ever saw the Hacksmith giant hexapod, but my first thought after seeing it (and seeing it be unsuccessful) is that an electric version would have been far better. Do you think it would be possible to build a human sized electric hexapod based on this design of yours?
5:11 Instead of making it snap on the vent holes maybe clip on around the leg or instead of clip, hook in the open areas on the leg and then use a pin and slide it through keyholes on the attached piece? 5:43 I mean like lower the plate and have it hook around and into the leg and then the build plate could have a lip so it could slide mount and there could be a dug out area like a button above the area the attachment locks in to keep it locked in.
you could do some electroplating for the armour it will look even more attractive and u could use magnets for the attachments so it will b easy for swapping different attachments
@@AecertRobotics You could scuff it up and cover it with graphite first, or use sputtering if you feel like building an apparatus for it. I think they make metallic spray paint too. Sometimes silver is used for optics, but I think you can get spray on copper from the auto parts store, it's sold as a gasket making material.
Are you planning to make a tutorial on how to program this robot? I watched the first video and really liked how the robot moved. If you're not planning to, could you recommend a tutorial to get started? I've been looking, but honestly, everything seems very confusing
You could add a microphone, send the audio to a server to have it voice recognised and converted into commands and coordinates and sent back. The phone may also be able to do it locally.
Also, you could modify one leg to have a further, retractable (foldable) and grasping element, so that upon need you lift that leg, you unfold the gripping element, and you can put objects on a basket placed on top at the centre of the pod. That retractable grasping element could be best controlled (both for unfolding and for grasping) using hydraulics (water or oil) so that you don't need to add an extra servo which would unbalance the pod, the 2 pistons would be in the center unit.
Both of these ideas are beyond the scope of this specific project, but i really really like them! I want to do both in the future, thanks for the suggestions 🙂
Have been playing Spiderman 2 recently. There is a spider bot in the game. Your design with custom add ons would be a perfect fit. Not sure if you would be interested to try it out. Will check If I can give it a shot in few months 😅
Where did you end up getting the cheaper 30kg servos? I've been trying to find some decently priced ones online, but they all seem to be $20+ per servo. The ~$12/servo ones you showed would be a great replacement for my current MG966Rs I'm currently using on my hexapod.
Unsure why screws were your only option? Surely some CA glue would do fine? I have huge prints held together with CA, and with a friction fit like that, it wouldn't go anywhere.
Could this sort of design be scaled up to carry any significant weight? something like that would be way out of reasonable budget for me of course.. but I was just thinking theoretically.. the strength of the servos are the main stumbling point I would think. that and the mahoosive battery needed. but to carry say.. the weight of a bag of shopping? or a large cat? lol. or is this just not where we are atm?
Excellent. What would I add to it... Hmmm. If you like complecation I would surchase an idea how about 3d printer on it back. Lol. I was looking at smallest printer possible and lightweight. Maybe not. Right now I am building robot dog and modify the 3d printers. I will build that Hexapod before I die.
Hi Aecert,I am faced with a problem I cannot figure out when I try to build my own hexapod. May I know how to determine if a battery is suitable for my hexapod project? I don't know how to calculate the current that servos will draw. Based on my understanding, a servo with a running current of 900mA and a stall current of 2.5A will draw 18 times the current, which is 16.2A to 45A. This seems unrealistic. I think my calculation is wrong, but I can't find any explanation.
Do your servos have a listed torque rating? In theory, the motor will stall when it's loaded with that amount of torque, but you shouldn't be getting even close to that during normal operation. The lifting servos at the base will have to work hardest, and the load will be pretty uniform there. But on the others, it really depends on what motion you have programmed. You might want to examine the effect of speed with an oscilloscope if you have one. Does it take less current to move slower? Is the output pulsed? Maybe a simple capacitor across the supply could take up the slack. As long as you don't have abrupt jerking movements at the base, they shouldn't be working that hard. There is some analysis you can do for a certain animation or motion, but in the end, there's no replacement for just measuring the current on the real thing. Hook it up to external power and run it through the motions with varying amounts of load.
Hello, thank you for releasing the project on your github. I am going to build a replica of your robot to learn trigonometry and make myself better in programming language C++. Thank you very much.
Hi I've been following you since the start, and I'm very impressed by your amazing work, keep going! I have a small question if you could help me with that can you give me the size of your robot when moving? like what's the max length and height it arrives to? cuz I should make a spider robot that can move in 50cm and I want to know if this is possible Thanks again
I think this could be scaled fairly easily. The hard part of this vehicle is the code and that's pretty much done. Since there are 100S of different servos available in many sizes, you could redesign one around the specs you need.
Hi, i see you got ACS71240KEXBLT-010B3-115 current sensor on a PCB, in documentation for it, it says that max voltage for it is 3.6v but on pcb it is connected to 5v. Is it correct? did you use leg switches in the code? and finally do you use 1 ubec for all servos? ds3225 can use up to 2.3A per servo, is 1 ubec capable of that kind of amperage? (for example if all 18 servos will work at max at the same time it can possible to draw up to 42A. Also when it comes to NiMH batteries they have a much worse dischare curve. In the hexapod i build before I connected servos directly to 2s lipo and even if documentation says max voltage is 6.8v those servos should be capable of withstanding 8.4V from fully charged 2s lipo.
Yup. The best way is to sense its voltage, and when it gets too low force it into some sort of sleep mode. What i currently do is just only run it for like 10 or so minutes.
You could create an ai and train it inside a simulation to adapt to different types of terrain (I think your current processor probably can't handle an ai so I recommend Raspberry Pi Or a broken phone)
@@AecertRobotics I use CA glue or superglue. CA is stronger but superglue is often good enough. What I tend to do is have a friction fit as you had done and then add glue, that way the alignment is perfect and the glue has to work less hard to maintain the bond
![How to make a REAL servo out of any motor. Multi-loop control. A robot dog? [Diary of a Madman №5]](http://i.ytimg.com/vi/yFH8key-t_M/mqdefault.jpg)
For permanent assemblies like the armour plates, I find that CA glue works miracles.
An FPV camera mount would be nice. Cheers
Glue. Wtf... Idk why i didnt think of that but yes that makes so much sense. Do you have a link to a preferred one?
@@AecertRobotics any brand of medium thickness CA. And get some "CA activator" it's a curing accelerator.
I was thinking the same thing lol, CA glue would have worked great for those parts
n @AecertRobotics i wouldnt use CA glue as it leaves smoke like effect on the parts thats not easy to clean. i use UHU all plast glue. it is great for plastic especially the printed parts
@@AecertRobotics i'd say poly cement is better, it fuses the plastics which is why its a modelmakers best friend
I have a few ideas
Add LiDAR and make it map its own environment
Add a nerf gun attachment and a button on the controller to shoot
Program it to be able to walk up and down stairs
#1 and 3 are much easier said than done!!!
#2 though.... stay tuned.
Wow already 13k subs! I remember you at 300 subs. Congrats!
Thank you so much, all this support has been amazing 😊
So happy to see you getting some good sponsors!
UA-camrs like this made me apply as an engineer to a uni, thx
Give it the body of a turret from the Portal game series. (Also loving the design so far, I know nothing of this sort of hobby, but this is quite enjoyable.
This project is top notch. I intend on making one. Keep up the good work !
Thank you so much, I really appreciate the support ❤️
A cool yet challenging attachment wuold be putting some electrode on the tip of the front legs, the one it uses for the slam attack, and wiring them with a high voltage booster so when it attack it can "zap" the target. you could also put some metal spikes on the already existing leg shields and do the same thing
2:55 I just heard kitten holder and i think i want a moving castle for my cats.
Some (very far fetched, but oh well) ideas: so if you’ve ever watched avatar 2, they have these swarm robots that build things and they kinda look a bit like this however they can walk up metal structures, could you add a small electromagnet on each foot? You can then power them when in contact with metal.
Another idea for holding things could be to add another servo to the central pair of legs to flip them 180 degrees so they face up. Then they could have slightly wider feet or grippers so they can pinch whatever they are grabbing together. Kind of like a human holding something over its head.
Some sensors to detect objects around would be cool especially partnered with 3D scan software to map or upload an environment.
Finally, the easiest idea, put a pen on the bottom and map its movement to draw something
I was searching some project that have a spider shape (Hexapod) to materialize an idea I had but not even in my best dreams I could imagine find something as awesome. Congratulations. I'm going to github to get all details about your project, thanks for that. If you have some tutorial or if you decide sell an kit about this project as you comment before please let me know. For a while I'll printing the parts shared by you and study all details about the assembly. My idea to this device is include a camera 180º to do some work at height on metalic structures using magnetic parts in the legs. I don't know exactly how do this but is something like that I have in my mind.
I think robotic arm on top will be excellent) Beautiful job!
you can easily screw directly into 3D printed part, just use self tapping screws, I've used this method on my 3D printed robot dog and it worked flawlessly
You can attach some devices on top like maybe a flying drone to get a different view of something and then when the hexapod sees or senses something and needs a different view to accurately understand it then it could. Maybe put a portable charger on it and a phone pad and then lay your phone on it and then youll never lose it because you can call it and bring your phone. Attach a speaker and mic that are small so the bot could understand speech and talk back like siri or connect an ai. A cam with a connection to ai would be cool too and help identify things and assist in work and finding things.
Great video and project overall!!
You probably could make the snap fit work if you rotate the orientation of the way you print the part. You can easily separate and snap across the layers of an FDM print, but if you change the orientation of the print so the bend direction goes along the "length of the filament" it gets a lot stronger.
I actually did try this, but the shape of the peg is extremely important, so it had to be printed right side up. Printing it sideways made it so the peg didnt fit properly into the hole. 100% right though it would stop the clips from snapping.
I think im going to go with glue for this sort of thing from now on.
@@AecertRobotics I also was going to say this (and I did pause the video so maybe its something you haven't addressed yet). If the problem with reorienting it is that the corners over bulge, thus ruining the fit you could try a small nozzle or only having one side be the "flexing" side, thus giving it a larger size to work with. But im assuming there is a reason the snaps are at the size you showed.
Honestly the screw holes in the armor look sick so besides making it a bit cheaper to reproduce the snaps don't seem that important. Mind if I ask where you get your heated inserts and screws from? I see plenty of options on amazon, but some have way too many variants to make it worth getting
I build a teaching tech hexapod ages ago. It worked ok but this one is faster stronger! Well done I'm learning a lot from your vids. Btw an arm that can grab things would be a great attachement. For something simpler maybe some distance sensors for autonomous walking. 😀
Oooh I'll have to check that out! Love the suggestions btw!
Wow. Love your creativity in this series. I really want to see more of it!
Depending on where you go with this project, a food dispenser would be hilarious! if I remember right, this robot is controlled by a person, perhaps a camera attachment could help you traverse easier. I'm thinking of a camera that can rotate 360, or by some margin, as a camera locked in place wouldn't be very flexible. I'm just blabbering really, awesome work on the robot, those attachments are looking mighty fine!
The form function in Fusion 360 will help you achieve those organic shapes like the armor plates much more easily. Also you'll find that for someone with a background in polygon modeling it's quite intuitive to work with.
buds got 13k subs and already getting some good sponsors keep up the good work
cool attachment idea, tho this might be as big a project as the hexapod itself, but with fusion 360 at your side, a lot of it may be easier than it sounds at first:
design a simple mechanism for firing airsoft pellets in full auto, could be as simple as a hopper feeding pellets thru a tube that becomes the barrel, with small gaps for spinning wheels to send the pellets flying, like hotwheels toys of yore, so that the barrel can be really short, allowing you to pack it into a housing that follows the existing portal design aesthetic of the leg armor, so it's like a walking portal turret, you could prolly control it's firing angle with a joystick and fire with a momentary switch
bonus points if you slap a camera on and make it fire at anything that moves, saying lines from the game thru a little speaker, like asking if anyone is there after it stops shooting, cause that would be hilarious🤣
I kinda had a bad feeling that you got bored during the process and gave up on the hexapod when it was unfinished. Nice to see you’re back in action
Edit: as for the ideas for attechments, i don’t know how much of a good idea this would be, since it won’t be functional, but maybe you can build an oval shaped head that carries replica infrared scanners and railgun and gatling gun mounts. The head can rotate x axis and the weapon mounts can rotate y axis, like a gimbal. This can make the hexapod look like a boss fight foe, that is out of a game with a cliche story where the hero fights with an evil genious, and has to battle the hexapod in his hideout to reach him.
imagine 1 cupholder and 1 arm to open the fridge and take a can
😂 It'd have to be a mini fridge obviously but holy crap that would be insane. A bit too complicated for this project but I love the idea!
this is better than guns
have a robot arm on the hexapod. With a robot arm (while more weight) you can instantely add a lot more functionality to the robot and shouldn't be so hard if you had already made a hexapod. With the robot arm you can make it grab anything or do anything with the right attachments to the arm.
Man this upgrades looks really cool.
I think putting propellers on the back of it would give it of a challenge and make the movement 3D. When you get the landings down you can try and use it as a flying hook.
I've been wanting to build a hexapod...this is awesome!
Add a big blow up hammer on top to hit stuff with and put a bit of amour on top so we can have battles with a sesneor which counts the amount of times hit
Dude this is unironically an amazing idea
still following your work. this is amazing stuff. :)
Thank you so much 🙂
looks amezing, especially with the armor plates.
my idea is stupid, over complicated and probably won't work... but, I would love to see a (single shot) nerf dart turret on top of this hexapod hahaha
Thank you!!
Hahaha not stupid at all, and you aren't the first to want this... 😄
Heated inserts and screws isn't the only way to make the armour rock solid, a decent epoxy (or hell even a decent CA glue) would probably be stronger than, or close to as strong as, the PLA
You are totally right, idk why i didnt think of glue tbh. Can you recommend a good one for pla+?
I should preface this as such:
I don't usually use PLA+, I usually print with normal PLA or sometimes nylon, as well as a lot of resin.
That being said, for CA glues even the cheapest stuff my local dollar store has works well (I've glued everything from ABS, PLA, resin, wood, my desk, and my fingers) so I don't have a specific brand recommendation per se -- if you live in NA (I don't anymore) gorilla glue should be easily available and fairly good, loctite super glue comes to mind as well and should be available almost anywhere.
As for 2-part epoxies, again even the cheap stuff usually works pretty well. But loctite springs to mind yet again as making pretty good epoxy glue.
3M makes a good CA glue as well, come to think of it.
Pretty much any brand known for making glues, for more specific recommendations I'd probably suggest you google around a bit for what works on your specific brand of PLA+ (formulations matter, and I don't use them on a day-to-day basis so I can't say for sure if I've ever glued any PLA+).
Or you could just print some test pieces and try different glues.
It's worth noting that some CA glues can take hours to fully harden, especially cheaper ones. I usually hold the part in place for about 10-20 seconds until the glue sticks it down good, then leave it overnight (about 24 hours or so, let's say) and it'll be rock solid.
Epoxies vary, but usually have longer open times and set times and I'd consult the instructions supplied with the specific glue.
it would be cool to add a turret and a camera
also, I would integrate the servo's power supply on the mega shield... By PCBway! cheers
how would i go about doing that?
@@AecertRobotics you have to add your own power regulator IC and filter caps. Look up buck converter circuits
I've been looking to find a good set of components to achieve this and at the price you paid, the BEC is actually a good choice. It would be challenging to achieve the same robustness at that price point. Also, it's the most likely part to fail in the circuit so having it external is not a bad idea. I have used the hextronic YEP controllers before and they are quite good.
"What is my purpose?" "You serve drinks."
😂
I want to see grippers as attachment!
Crab like?
@AecertRobotics this project is amazing. i was following it from the start. I ended up building it. the build went smoothly, got all the parts and electronic, connected everything. there was no smoke luckily. unfortunately, i was unable to get pass the calibration state. one servo ended up dying. fried completely from inside. I only need help in this stage if anyone can help please. just P.S. i didnt add the current sensor. there was nothing in the code that suggests that it was in use. I just bridged the too current pads to override it.
You could add an aimable light or something, I imagine you'd need more software though.
Instead of a bec you could try a linear regulator on the pcb to get the voltage for the servos
I will look into this, thank you!
@@AecertRobotics i would recommend the lm317 becuase you can pick any voltage you like under 20v if you just want 6v you can use the lm7806
i would love to see ai controlled so you can talk to it and tell it to do stuff. great video
I really love this idea. It's out of the scope of this project since it would require way more processing power, but if and when I ever make another hexapod with more advanced electronics, I will 100% be doing this.
one idea i'd love to see mounted on the hexapod is a cannon with a cammera on it that would allow you to control the robot, aim to a target through the camera and shoot the target, let's say a can of soda. it would be awesome !
Hears an attachment idea , a standard box that can be configured for cargo , a sensor suite and or additional computer function.
or it would be cool if it could have a top mount for a mini projector so that it can project pattern on the ceiling or even just a ring of lights that can express things like when the battery is low, when it's starting an attack or charging (if you are gonig to make him rechargable without disassembling). Obviously the projector could make all this things other. Sorry the spam but right now i can't realize them on my own so seeing them wuold be really cool
Dont apologize, keep the ideas coming!! 🙂
If you want more inspiration look at Mark Setrakian history of robots, specially Mechadon, a 500lbs battle bot he rode in on. Though this battle bot wasn’t ideal for battles, it was cool. For cooler ones look to The Animatronic Robot Designs of Mark Setrakian in Adam Savage’s Tested. See the hand gently, though fluidly spinning a globe. I think your design is really cool and also reminiscent of his designs
love the vid. I remember you mentioned a full course including coding in the first video, really looking forward to it. Care to share some details about the course?
When robot engineer says "Hold my beer" ...
🤣
This is soo good!
Thanks 🙂 I really appreciate the support!
Probably some set of sensors for surroundings mapping and trying to learn hex how to navigate in space and time. It‘s pretty interesting coding challenge as well
Another UA-camr, the 3dprintingprofessor created a good snap fit connector for his printablok project. It would be cool to see it used on your hexapod.
I may have seen it but im not sure. Can you link me the video?
@@AecertRobotics https ://ua-cam.com/video/BU7ZYPHWDo4/v-deo.html
Nice video, well done, thank you for sharing it with us :)
Any ideas for the weight limit?
It would be cool to see a cat mount
Love this series, been super interesting to watch. I don't know if you ever saw the Hacksmith giant hexapod, but my first thought after seeing it (and seeing it be unsuccessful) is that an electric version would have been far better. Do you think it would be possible to build a human sized electric hexapod based on this design of yours?
It's definitely possible!! Just very expensive 😅
I started looking into proportional hydraulic valves with this very idea in mind.. 18 valves would cost the price of a new car.. lol
The spider bot is fast !
If it was any faster you could race it .
Some sort of time trial obstacle course would be so cool!
if you already have armor make weapons to put on the roll cage. Or 2 arms that can grab things although the weight would be a challenge.
5:11 Instead of making it snap on the vent holes maybe clip on around the leg or instead of clip, hook in the open areas on the leg and then use a pin and slide it through keyholes on the attached piece?
5:43 I mean like lower the plate and have it hook around and into the leg and then the build plate could have a lip so it could slide mount and there could be a dug out area like a button above the area the attachment locks in to keep it locked in.
you could do some electroplating for the armour it will look even more attractive and u could use magnets for the attachments so it will b easy for swapping different attachments
i need to be careful with magnets since they can fall off from the hexapod walking. Can you electroplate 3d prints? That sounds amazing if you can!
@@AecertRobotics You could scuff it up and cover it with graphite first, or use sputtering if you feel like building an apparatus for it. I think they make metallic spray paint too. Sometimes silver is used for optics, but I think you can get spray on copper from the auto parts store, it's sold as a gasket making material.
Cool video!
Are you planning to make a tutorial on how to program this robot? I watched the first video and really liked how the robot moved. If you're not planning to, could you recommend a tutorial to get started? I've been looking, but honestly, everything seems very confusing
attachment idea add a laser light to play with the cat
You could add a microphone, send the audio to a server to have it voice recognised and converted into commands and coordinates and sent back. The phone may also be able to do it locally.
Also, you could modify one leg to have a further, retractable (foldable) and grasping element, so that upon need you lift that leg, you unfold the gripping element, and you can put objects on a basket placed on top at the centre of the pod. That retractable grasping element could be best controlled (both for unfolding and for grasping) using hydraulics (water or oil) so that you don't need to add an extra servo which would unbalance the pod, the 2 pistons would be in the center unit.
Both of these ideas are beyond the scope of this specific project, but i really really like them! I want to do both in the future, thanks for the suggestions 🙂
Have been playing Spiderman 2 recently. There is a spider bot in the game. Your design with custom add ons would be a perfect fit.
Not sure if you would be interested to try it out.
Will check If I can give it a shot in few months 😅
A nerf turret would be cool also a camera and video so you can fpv it 😂😂 this thing is so cool cant wait to build one
You better make a laser turret attatchment on the top
Where did you end up getting the cheaper 30kg servos? I've been trying to find some decently priced ones online, but they all seem to be $20+ per servo. The ~$12/servo ones you showed would be a great replacement for my current MG966Rs I'm currently using on my hexapod.
Great question!
www.aliexpress.us/item/2251832533583005.html?spm=a2g0o.order_list.order_list_main.10.3c871802fLZ94i&gatewayAdapt=glo2usa
@@AecertRobotics , I believe, you had 35kg servos before, correct? So they were too much and you decided to downgrade all servos to 30kg ones?
Correct. I downgraded purely because of cost!
you should make plates that can turn into a ball
the legs arent designed to bend in a way that would let me do that, but i love the idea!
@@AecertRobotics awww alr
Unsure why screws were your only option? Surely some CA glue would do fine? I have huge prints held together with CA, and with a friction fit like that, it wouldn't go anywhere.
Could this sort of design be scaled up to carry any significant weight? something like that would be way out of reasonable budget for me of course.. but I was just thinking theoretically.. the strength of the servos are the main stumbling point I would think. that and the mahoosive battery needed. but to carry say.. the weight of a bag of shopping? or a large cat? lol. or is this just not where we are atm?
How about a blade mounted in roughly the spot of the leg armour? Stabby hexapod?
Excellent. What would I add to it... Hmmm. If you like complecation I would surchase an idea how about 3d printer on it back. Lol. I was looking at smallest printer possible and lightweight. Maybe not. Right now I am building robot dog and modify the 3d printers. I will build that Hexapod before I die.
A nerf turrets could be cool
When the servo board will be available? By the way the armor looks awesome.
a nerf turret that shoot flies would be nice.
Definitely won't be done before the release, but I 100% plan on adding some sort of turret.
It's usually a bad idea to design injection molding pegs for 3D printing.
What do you mean?
add a hand that does 👌 on command
😂
@@AecertRobotics and maybe tasers too 🤔
Water gun attachment so it shoots soda into your mouth
add lasers
what about a small airsoft turret?
Maybe not airsoft... but i love the idea!
is that shark on the background , can we have a look ??!!
It is!! Ummm whats the best way i can show you lol? maybe an outro in a video?
LEDs!
reprint your clamps just rotate them 90 degrees and they will be stronger.
1 idea you can make a servo arm on the hexa pod
Hi Aecert,I am faced with a problem I cannot figure out when I try to build my own hexapod. May I know how to determine if a battery is suitable for my hexapod project? I don't know how to calculate the current that servos will draw. Based on my understanding, a servo with a running current of 900mA and a stall current of 2.5A will draw 18 times the current, which is 16.2A to 45A. This seems unrealistic. I think my calculation is wrong, but I can't find any explanation.
Do your servos have a listed torque rating? In theory, the motor will stall when it's loaded with that amount of torque, but you shouldn't be getting even close to that during normal operation. The lifting servos at the base will have to work hardest, and the load will be pretty uniform there. But on the others, it really depends on what motion you have programmed. You might want to examine the effect of speed with an oscilloscope if you have one. Does it take less current to move slower? Is the output pulsed? Maybe a simple capacitor across the supply could take up the slack. As long as you don't have abrupt jerking movements at the base, they shouldn't be working that hard. There is some analysis you can do for a certain animation or motion, but in the end, there's no replacement for just measuring the current on the real thing. Hook it up to external power and run it through the motions with varying amounts of load.
Hello,
thank you for releasing the project on your github.
I am going to build a replica of your robot to learn trigonometry and make myself better in programming language C++.
Thank you very much.
Good luck! I would highly recommend waiting for the kit before trying to build it!
@AecertRobotics
Oh.
So. Do you want to sell the robot in a kit?
What will be included in the final kit?
Yup, so everything that you need will be included except the 3d printed parts and the tools (screw driver, soldering iron, 3d printer, etc)
@@AecertRobotics do you have any idea on what time the kit will come out?
A nerf gun attachment!
Awesome.
Hi
I've been following you since the start, and I'm very impressed by your amazing work, keep going!
I have a small question if you could help me with that
can you give me the size of your robot when moving? like what's the max length and height it arrives to?
cuz I should make a spider robot that can move in 50cm and I want to know if this is possible
Thanks again
I think this could be scaled fairly easily. The hard part of this vehicle is the code and that's pretty much done. Since there are 100S of different servos available in many sizes, you could redesign one around the specs you need.
make an extension attachment that can fire a nerf gun
Hi, i see you got ACS71240KEXBLT-010B3-115 current sensor on a PCB, in documentation for it, it says that max voltage for it is 3.6v but on pcb it is connected to 5v. Is it correct? did you use leg switches in the code? and finally do you use 1 ubec for all servos? ds3225 can use up to 2.3A per servo, is 1 ubec capable of that kind of amperage? (for example if all 18 servos will work at max at the same time it can possible to draw up to 42A. Also when it comes to NiMH batteries they have a much worse dischare curve. In the hexapod i build before I connected servos directly to 2s lipo and even if documentation says max voltage is 6.8v those servos should be capable of withstanding 8.4V from fully charged 2s lipo.
make the hexapod shoot people in the chest with a airsoft/paintball gun
is there any way to prevent the battery from over discharge
Yup. The best way is to sense its voltage, and when it gets too low force it into some sort of sleep mode. What i currently do is just only run it for like 10 or so minutes.
You could create an ai and train it inside a simulation to adapt to different types of terrain (I think your current processor probably can't handle an ai so I recommend Raspberry Pi Or a broken phone)
That is muchhhh easier said than done 😅
If you are making a piece that never needs to be disassembled, like the armour, why not just glue it?
Idk.... it just never crossed my mind. Do you have any recommendations for gluing pla+?
@@AecertRobotics I use CA glue or superglue. CA is stronger but superglue is often good enough. What I tend to do is have a friction fit as you had done and then add glue, that way the alignment is perfect and the glue has to work less hard to maintain the bond
camera attachment and then talk nice to insta360 for one of those stabilized cameras I'm seeing all over engineering youtube.
How i can print one of these creatures myself?
I wish I knew how to make PCBs
can you make a video of the fusion 360 3 hour tutorial but shorter
nerf (or bb) turret?
please add a nerf style blaser to it
What are you using for a tip of the leg?
nerf gun turret attachment, or something similar lol