I have started the research into the Solo-12 robot and thought that somewhere, but cannot remember now where, someone saying that it might be able to carry 500g or 1 Kg.
What I am interested in is how to perhaps upgrade to another motor type so that it could carry some rechargeable batteries and some additional components.
Right now, it’s an investigation but something that is probably needed since the current designs seem to have the Solo-12 on a power/communication tether.
Any thoughts or suggestions?
Best Regards and have a good day.
Forgot to mention that what I am also trying to get a feel for as well is the total power consumption to try and size up and estimate a battery pack and BMS for this system. My understanding is that the MIT Mini Cheetah uses a Kobalt rechargeable battery for their robot which gives about 90 minutes of run time.
we haven’t done any payload testing with the robots yet.
I would think that the Solo12 robot could carry 500g-1000g depending on the gait, the leg posture and the task.
The robot was designed for high dynamic performance - not for carrying high payloads.
To improve the load carrying capabilities you could use a motor with higher torque output or increase the gear reduction. Improving the cooling would also help to stay below the thermal limits of the motors.
We are currently working on integrating onboard batteries, imu state estimation and wireless communication. For that we are developing a power management board that will monitor the batteries and prevent damage.
We are planning to have two batteries in series - one in the front and one in the back. (picture below)
We’ll start with this 850mAh LiPo battery and will evaluate how much run-time we can get from it.
Pierre-Alexandre had shared data logs of the robot performing a trot (on the power supply).
You can find the topic and a link to the data here.
I am getting ready to try and build the Solo-12 (modified) and am currently sourcing some stronger motors with a bit more torque as you have mentioned so as not to degrade dynamic performance and allow for greater lifting capacity since my goals are to also include a mini-computer with wireless/Bluetooth capabilities as well but may have to scale the Solo base design a bit to make that happen.
Additionally, I want to add stereo cameras and a few other nice features if possible going forward.
Of course, I am definitely interested in having some onboard batteries while am actually also investigating the idea of inductive wireless charging while also having a charging port that can be plugged into from an external source.
For the batteries themselves, I wanted to see if there are any reasonable off-the-shelf batteries that could simple be plugged into the robot like in cordless power tools to keep it simple and use commodity components.
What is the current status of this? I’m sure many people want to use onboard batteries, and as was mentioned earlier, this requires a power-management board.
However, its main component ( a stm32F103c8t6 MCU ) is almost impossible to get these days, worse, many fake chips which are not 100% compatible exist and are sold instead of the real deal.
Feel free to replicate this board, but pay attention to the chipset you solder on, as I was not able to make it work with non-genuine STM parts.
As this board as not yet been adopted by any user we know of, we are wondering if redesigning the board around another MCU won’t be a better solution (like a widely available MCU such as RP2040 or ESP32).
Because of this I can not guaranty that the board you start assembling now will be the standard in a few months.
In the same time, I don’t have the bandwidth to redesign this board in the short term, and can not tell when / if it will be the case.
I know this might by frustrating,
Another solution is to use an off the shelf board like mjbots power dist board overview - YouTube to safely connect your battery to the robot electronic. Note that this particular board won’t communicate with the masterbord, as we don’t use the same communication buses.
Thanks for the comprehensive reply. I’ll hold off until the situation becomes clearer, in the hope that someone will either identify a chip which is really 100% compatible, or come up with some other solution. I’m not competent to redesign the board around a different chip myself.
If nothing else comes along I’ll look at the mjbots board and how difficult it would be to integrate it.
Being able to run SOLO 12 without a “leash” / tether via an onboard power supply, onboard compute (e.g. a Raspberry Pi 4) and perhaps some onboard sensors would be a great option indeed.
Any chance to buy a powerboard (or entire autonomy kit) somewhere? Ideally assembled (PCBA), as a kit (PCB + components) or at least the PCB?
(Doing single quantity PCB runs and sourcing the components for a single one is always such a pain.)
Hi! I looked over this page and wondered if you built your Solo12 with modifications. I am looking for the same goals as you were, wondering what motors you used, and the modifications you had to make to the design. Thanks for the help!
