How can I evaluate the carrying capacity of a quadruped/hexapod robot?

[theycallmevirgo]

TL;DR

How many legs must defeat overall mg at the same time to allow for motion?

How can I predict carrying capacity of something like this ?

or this?

Clearly some combination of servos must defeat mg, but how many? Do all of them need to be powered constantly, whether any leg is moving or not, just to "stand up straight"?

Further, each leg is clearly a lever. What is the design equation to maximize vertical force?Many thanks in advance for any help

Joe

 

[Baluncore]

There is no single equation that will answer your question. You will need to define a geometric model of the mechanical linkage. Motor force will be transmitted through those links to become a vertical force on the ground.

You will also need to specify how the mass load is distributed over the vehicle chassis.

The number of degrees of freedom in each leg will decide how many feet may remain on the ground while walking. So there may be several modes of operation, some of which may lead to balance or stability problems. Taking shorter steps may support greater loads.

Can you draw a diagram of the linkage that defines one leg?

 

[theycallmevirgo]

Sorry if I have to do this piecemeal, but I may lose network soon. To answer one of your questions immediately, safe to assume a fairly even horizontal load distribution. Maybe slightly more in the center.

I can easily do a fairly detailed diagram of the linkage. Do you prefer I make a separate reply or edit this one?

 

[DaveC426913]

I'm not convinced that blue contraption has any capability of controlled leg movement.

I see only two arms per leg, both attaching to a single axle (i.e. mechanically, it's just one arm with a redundant arm), so Ill bet if it really existed, it would just fall flat.Contrast with the wooden one, which correctly has two arms, each with its own attachment point.

Never mind. I just noticed that Blue's legs have a motors that their knees.

 

[theycallmevirgo]

I'm not convinced that blue contraption has any capability of controlled leg movement.

I see only two arms per leg, both attaching to a single axle (i.e. mechanically, it's just one arm with a redundant arm), so Ill bet if it really existed, it would just fall flat.Contrast with the wooden one, which correctly has two arms, each with its own attachment point.

Never mind. I just noticed that Blue's legs have a motors that their knees.

They all have motors in their knees.

 

[Baluncore]

I can easily do a fairly detailed diagram of the linkage. Do you prefer I make a separate reply or edit this one?

Keep the discussion in this thread. Post your diagram here.

We need to know the length of links between pins and from pins to the tips of the feet. Do you have torque specifications for the actuator servos ?

 

[FEAnalyst]

I have a book about walking robots and it features various design calculations (including evaluation of necessary torque and power) but it's in Polish, unfortunately. However, there are some articles and boook chapters that can be useful for you. For example check the "Design of Hexapod Walking Robots: Background and Challenges" chapter of the book titled "Handbook of Research on Advancements in Robotics and Mechatronics".

 

[theycallmevirgo]

I have a book about walking robots and it features various design calculations (including evaluation of necessary torque and power) but it's in Polish, unfortunately. However, there are some articles and boook chapters that can be useful for you. For example check the "Design of Hexapod Walking Robots: Background and Challenges" chapter of the book titled "Handbook of Research on Advancements in Robotics and Mechatronics".

Thanks, will google

 

[theycallmevirgo]

I have a book about walking robots and it features various design calculations (including evaluation of necessary torque and power) but it's in Polish, unfortunately. However, there are some articles and boook chapters that can be useful for you. For example check the "Design of Hexapod Walking Robots: Background and Challenges" chapter of the book titled "Handbook of Research on Advancements in Robotics and Mechatronics".

Actually I am (fluent in) Russian so I may possibly be able to make my way through polish. Is it softcopy?

 

[FEAnalyst]

Actually I am (fluent in) Russian so I may possibly be able to make my way through polish. Is it softcopy?

The title of this book is „Maszyny kroczące” and it’s written by Teresa Zielińska. You can buy e-book version here:
https://ksiegarnia.pwn.pl/Maszyny-k...rowanie-i-wzorce-biologiczne,824176657,p.html

Let me know if you need any help with translation.

 

[Baluncore]

I found two books in English, on the subject of walking robots, edited Teresa Zielińska.

Romansy 16: Robot Design, Dynamics, and Control [1 ed.]
Atsuo Takanishi (auth.), Teresa Zielińska, Cezary Zieliński (eds.)
CISM Courses and Lectures 487
ISBN: 978-3-211-36064-4, 978-3-211-38927-0, 3-211-36064-6
Springer-Verlag Wien 2006 [463 pages] English

Walking: Biological and Technological Aspects [1 ed.]
Friedrich Pfeiffer, Teresa Zielinska (eds.)
International Centre for Mechanical Sciences 467
ISBN: 978-3-211-22134-1, 978-3-7091-2772-8
Springer-Verlag Wien 2004 [194 pages] English

 

[Lnewqban]

As you can see from the picture, the vertical position and movement of each leg is controlled by one servo, which arm is anchored to the lower linkage.
If you gradually add weight to the chassis of the robot, a point will be reached at which the stall torque value of the four servos will stop further control and movement of the legs.

Please, see:
https://www.societyofrobots.com/actuators_servos.shtml#digitalanalogservos

Common servos don't have any self-locking mechanism, reason for which, the motors of these four servos will be using as many amps as needed to compensate for the required torque to keep each leg at certain height (at least while walking).

That is not the case for the four servos controling the radial position and movement of the legs, but increasing weight on the chassis will also increase friction on those horizontal linkages, requiring additional output torques and input Amps.

If walking on non-flat surfaces with extra weight loaded chassis, the strength and rigidity of the legs and linkages under lateral loads should be considered.

 

[theycallmevirgo]

As you can see from the picture, the vertical position and movement of each leg is controlled by one servo, which arm is anchored to the lower linkage.
If you gradually add weight to the chassis of the robot, a point will be reached at which the stall torque value of the four servos will stop further control and movement of the legs.

Please, see:
https://www.societyofrobots.com/actuators_servos.shtml#digitalanalogservos

Common servos don't have any self-locking mechanism, reason for which, the motors of these four servos will be using as many amps as needed to compensate for the required torque to keep each leg at certain height (at least while walking).

That is not the case for the four servos controling the radial position and movement of the legs, but increasing weight on the chassis will also increase friction on those horizontal linkages, requiring additional output torques and input Amps.

If walking on non-flat surfaces with extra weight loaded chassis, the strength and rigidity of the legs and linkages under lateral loads should be considered.

Thanks so much for this, I think I will test experimentally based on your suggestions and follow up with any additional questions.

 

[Lnewqban]

Thanks so much for this, I think I will test experimentally based on your suggestions and follow up with any additional questions.

You are welcome.
Best luck with the project.