Limbi: Preliminary Design

Project Objective/ Mission Profile

By: Alondra Vivas

Project Objective

We plan to continue the advancement of NASA JPL’s Limbi, a multi-jointed robot that allows for the connection of two modules through their docking mechanisms. The goal is to improve on the current Limbi by developing a mechanical androgynous connector for Limbi as well as the cubic modules, reduce the overall size of NASA JPL’s Limbi and mass, use renewable energy to allow Limbi and the cubic modules to produce power, allow power transfer to flow from the cubic modules to Limbi and between each other, and using a low friction surface to replicate space as close as we can while reducing the cost of the overall project.

Mission Profile

The purpose of Limbi is for astronauts to create structures autonomously in space that would reduce the time, risk, and cost of building the structures. Utilizing the idea of Legos, the cubic modules can be used to build space stations, large solar panels, and space laboratories for safer and more efficient experiments.This would be accomplished by having Limbi pick up one cubic module with its connector, attach it to another cubic module while remaining attached to the second cubic module,the next step would be to pick up another cubic module and repeat the process over again. 

Project Requirements

By: Alondra Vivas

The second generation Limbi follows the Level 1 requirements of the first generation Limbi with a few adjustments, to compare see Limbi geration 1 requirements. These requirements must be accomplished while also taking in to consideration that this project is meant to be used in space. 

Engineering Standards and Constraints

  1. The Limbi shall employ a custom PCB to extend the functions of the Arduino Leonard by allowing control of 4 GM9 motors, and logic levels.
  2. Disassemble and Reassemble of the robot shall be constrained to less than 20 minutes (10 minutes+10 minutes).
  3. The Limbi shall be completed by the date of the final: December 17th 2019.
  4. The robot shall be designed in such a way that there are no dangling or exposed wires via cable wrap. 
  5. The form factor of Limbi shall be constrained by the original JPL version.
  6. The usability of the Limbi shall be enhanced by use of the Arxterra phone and control panel application
  7. The ArxRobot app shall allow control of all joint servos and docking servos (see requirement L 1.10)
  8. Back of the envelope calculations and experiments shall be conducted to set the diameter of power carrying wires. Follow the American Wire Gauge (AWG) standard when defining the diameter of power carrying wires.
  9. Manufacturability of 3D printed robots shall minimize the number of files to be printed when using the library’s Innovation Space to print the final robot (waived)
  10. All Lithium (Li-ion, Li-polymer) batteries shall be stored, when not in use, in a fire and explosion proof battery bag.
  11. Software shall be written in the Arduino De facto Standard scripting language and/or using the GCC C++ programming language, which is implements the ISO C++ standard (ISO/IEC 14882:1998) published in 1998, and the 2011 and 2014 revisions.
  12. The Limbi shall be controlled via Bluetooth 4.0 in compliance with the Bluetooth Special Interest Group (SIG) Standard (supersedes IEEE 802.15.1).

Level 1 Program and Project Level Requirements

  1. The objective operation of a  low friction surface will be implemented. 
  2. The Limbi project will be smaller than the JPL Limbi for cost and storage purposes.
  3. The arm’s 4 joints shall move each of their attached limbs accordingly, with 4 of the 5 limbs moving in total and the final limb acting as a stabilizer.
  4. Each joint shall have 180 degrees of movement in one plane (x-y).
  5. The arm shall be able to connect and disconnect with the cubic modules.
  6. Docking mechanism shall keep the arm and module connected as the arm moves until it is meant to be disengaged.
  7. The arm shall have a docking mechanism on each end to connect to two modules at once.
  8. The arm shall be able to move the module (in the same plane as the actuator planar scope).
  9. The arm shall be able to lock two modules together through a mechanical design.
  10. The movement of the arm shall be controlled by the user with custom software.
  11. The Limbi shall be controlled with a micro-controller (3dot)
  12. Each cubic module shall have the capability of providing power to the arm and to each other cubic modules.
  13. The arm and limbi shall only be powered with one rechargeable battery and solar cells. 
  14. For demonstration purposes the module shall have docks on all 6 of its faces.
  15. The module should indicate when a secure connection is made between the Limbi and modules with an LED.
  16.  One module shall be defined as the base module and shall be stationary to represent a large, unmoving mass in space (such as the spacecraft). This requirement is based on Section 4 of “An Untethered Mobile Limb for Modular In-Space Assembly”.    

Level 2 Program and Project Level Requirements

  1. The arm will be supported with 8 small nylon metallic ball casters (each weighing 1.1g) on Limb 1 and Limb 3 to simulate the conditions where the arm will not be affected by gravity.
  2. The Limbi will follow the form factor of the JPL version; the lengths will be optimized in respect to inverse kinematics.
  3. Each joint shall control 2 limbs at a time.
    1.  Joint 1 shall control the motion between Limb 0 and Limb 1.
    2. Joint 2 shall control the motion of between Limb 1 and Limb 2.
    3. Joint 3 shall control the motion between Limb 2 and Limb 3.
    4. Joint 4 shall control the movement between Limb 3 and Limb 4.
  4. Limbi will have 4 joints controlled by GM9 motors.
    1.  Each GM9 motor shall require no more than 6V to run with a load.
    2. The GM9 motor shall be able to provide more than 0.539 Nm or 5.5kg/cm based on the force to move an object in a planar field.
  5. The docking mechanism shall consist of the interlocking mechanical device described allows the Limbi arm to successfully attach and detach to and from the module.
    1. The arm shall be able to attach and detach to and from a module without pushing the module away due to electromagnets.
    2. The module will have an androgynous connector on all 6 faces.
  6. Only 1 docking DC motor shall be in motion at once so the module does not un-dock while the other module docks (this would cause power loss).
  7. The docking mechanisms on each end will be identical to each other.
  8. The connector is 7cm X 7cm X 1.25cm.
  9. The connector’s weight is approximately 10g.
  10. The custom software will be implemented through the Arxterra App.
  11. The user interface shall utilize wireless control of Limbi.
  12. The user interface shall have push buttons/toggles for pre-determined movements.
  13. The user interface shall use simulated sliders for manual movements.
  14. The project will use a 3dot board mounted within the Limbi arm to allow the servos to be controlled directly by the microcontroller.
  15. The power provided from the module shall come from a battery and/or the solar cell via the docking mechanism.
  16. The battery shall be capable of providing 1055 mA current (if needed) which will be used to power the MCU, 6 servos, Bluetooth Module, TFT LCD Display, and TTL Serial Camera.
  17. The battery used will be a rechargeable AA battery weighing 22g each and using 1.5V each. 
  18. All of the 6 faces on the cubic module will be both f0r interconnections among each and to connect to Limbi.
  19. The LED on the module should be activated by one of the four power connections.
  20. The LED should be on the corner so the person controlling the app can easily see that a secure connection has been made

Project Breakdown Structure

Mechanical Design


By: Edward Villanueva

We have separate ideas regarding the androgynous connectors. There are several requirements that are needed within the connectors, to ensure that modules can be docked to limbi and one another, with minimal difficulties.

Module Requirements:

  1. The connector needs to have a mechanical lock, so that a constant power source is not required to stay locked into place.
  2. The lock must be rigid so that long chains of blocks stay stable and do not fluctuate.
  3. The connection must have a soft-docking feature, meaning excessive force is not required to connect the two modules together, as this would prove troublesome in space, where there is no friction to stop a floating module.

Androgynous connector ideas:

  1. Snap rivets in conjunction with Velcro, which would allow for a simple push to lock the connectors in place.
  2. A fire hose type connection, in which two androgynous connections are placed facing one another, and both twisted clockwise to create a nearly inseparable bond between the two connectors.
  3. A bottle cap connection, in which one connector screws into the other, creating a bond that could only be unscrewed to separate.
  4. A screw method that would see 2 screws and two threaded holes per box, and when coming together, the screws would be screwed into the threaded holes to create the bond.


By: Alondra Vivas

The first generation Limbi used servos in each of the joints. The problem with using servos was that it added more overall mass to Limbi, was tougher to work with, and would create an unbalance where the servo could not hold a limb in place. Analyzing these issues, we think the best approach for the motors would be to use GM9 motors on each of the joints (4 in total). It also has a stall torque of 5,500g*cm, 78RMP, and has a mass of 31.4g each. Therefore, the benefit of the GM9 is in its smaller size, less mass, and is simpler to control in comparison to the servo. 

Each connector must be mechanically designed in order so that they don’t need power to flow through them to lock or unlock the connectors. Using this idea, the motor we are further analyzing is the toy DC motor. It has just the right amount of torque we would need to rotate the connectors and join two modules together. Coreless motors are also being looked into, due to their small size and high rpm.

In order for the motor to work on Earth as it would in space we would need a friction less surface. The friction coefficient in space is 0. On Earth it is extremely difficult to achieve this friction coefficient. 

Types of friction less surfaces:

  1. Ice has a friction coefficient between 0.04-0.05. An issue with using ice as our friction less surface is in that we would have to take Limbi to the Ice Rink each time we need to test it which would result in schedule conflicts.
  2. Hockey tables have a friction coefficient of 0.01-0.03. The tests to figure out the friction coefficient of a hockey table have been tested when there is a puck on the table. We are hesitant to use the hockey table because it would be cost inefficient, would take a considerable amount to be built, and it would be problematic to transport.
  3. Dry sprays have a coefficient of friction of 0.04-0.05, making them about as friction less as ice. The added convenience is the spray could be placed on any surface to reduce the overall friction, and the surface, such as a fold table table could be taken with us very easily.  

Limbi’s Design

By: Edward Villanueva

Limbi’s design will be very similar of JPLs limbi, borrowing many ideas from the original and improving upon JPL’s major points of failure.

Major design points of Limbi:

  1. The size will be reduced to about 45% of the original limbi, with its total length of the limbs being about 70 cm in length.
  2. The proportions that JPL calculated for the limbs will be kept, as a program was ran to create the proportions that would allow limbi to perform its tasks.
  3. The ball casters on the modules and limbi itself will also be kept to allow for smooth movement along a nearly frictionless surface.
  4. The design of limbi will also be reminiscent of JPL’s where limbs 1 and 3 will be below the others to allow the connectors on the end of limbi to line up with the connectors on the modules.
  5. The overall mass of limbi will be reduced by approximately half of the first generation Limbi, allowing for lighter motors and easier movement of the limbs.

Inverse Kinematics

By: Edward Villanueva

The inverse kinematics of Limbi refers to the usage of kinematics equations to figure out the proper angles that are needed for Limbi to reach the desired location. The difference between forward and inverse kinematics lies in the fact that in Forward kinematics the joint angles are already known, and the end point of the final limb is what is being calculated. However in inverse kinematics, the final limb position is known, and the joint angles are what is being calculated.

In our case, the calculations of limbi’s motors will be done after manually controlling limbi to achieve the desired position, and programming a preset of animations in which limbi will automatically grab a module and place it in the desired position. However, this is not true inverse kinematics as cubic modules will be placed in preset locations in order for limbi to be able to connect modules together. The final version of limbi will likely have sensors to scan the distance and angle between itself and the module it would like to connect to, and then take measurements and perform the necessary calculations to autonomously attach to this module.

Electronic System Design

Arduino and 3Dot Board

By: Colin Rogers

The 3Dot is the “brain” of Limbi.  It will be responsible for controlling all aspects of the unit.  The limbs, electromagnets, and connectors will all be controlled by a GUI on a remote computer via Bluetooth.  This allows the user to communicate with the 3Dot directly and in turn control the Limbi. The cubes also need to be arduino controlled. This allows all the connectors on the cubes to be individually controlled.  Cubes will be able to detach from one another without the aid of Limbi.

Important Points

  1. 3Dot will be used for Limbi
  2. Will be controlled by a remote computer
  3. Cubes will be arduino controlled
  4. Cubes will be controlled independently of Limbi


By: Colin Rogers

The wires in Limbi  will take a detachable approach.  This means that instead of many solder joints, splice connectors will be used.  This ensures that Limbi can be easily assembled and disassembled when needed.  The wiring in Limbi will be mostly hidden.  No long wires will be seen from the outside because the wiring will be concealed from view. This gives Limbi a very neat and finished look.

Important Points

  1. Most of Limbi’s wiring will be detachable 
  2. Splice connectors will be used in place of many solder joints
  3. Wiring will be hidden as much as possible


Renewable Energy

By: Alondra Vivas

Limbi will be working in space, therefore it will be using solar energy to charge the renewable batteries we will be putting in place. We plan on using lightweight solar cells, each weighing 19g. It’s position will be placed on L2 (the middle limb) on the top and bottom , by placing it only on two sides we are reducing the mass and allowing the sun (even if it’s only partial) to hit at one solar cell at all times unless something blocks Limbi or the structure. We also plan on using the solar cells on each cubic module and to place it on the top and bottom as discussed previously. 

The circuit design we will be using is by David Cook, the website creator of RobotRoom: Solar Recharging Circuit. He demonstrates a schematic of a solar panel charging circuit with voltage measurements, which is also shown below. 

Schematic of solar panel charger circuit with voltage measurement.

Power Transfer

By: Colin Rogers

Limbi will have a power transfer feature.  This essentially means that the cubes will be capable of transferring power to one another as well as to Limbi.  This is to assure that all the components have the appropriate amount of power that they require.  Modular contacts will be used in the connectors of the cubes and Limbi in order to easily pass power between devices.

Design and Unique Tasks


The first generation Limbi will be duplicated in the second generation Limbi but with some changes. As stated in the paragraph titled Limbi, we will be reducing its size, mass, and material it gets printed on. However, we will implement the same amount of limbs, joints, and connectors to Limbi. 

Rapid Prototyping

The second generation Limbi should function similarly to the first generation Limbi. 3D printing has been in motion since mid August. We should accomplish the prototype by the beginning of October, meaning we will buy all the necessary materials and finish 3D printing by September 20th, 2019.