Soft robotics arm

Inspiration

Soft robotics field has been growing rapidly as the soft nature of these robots allows them to complete tasks that traditional rigid robots cannot do. Especially in the field of clinical operations, soft robotics significantly reduces chance of damaging its surroundings, which allows physicians and doctors to perform safer operations. With interests in clinical applications of soft robotics, we decided to focus on developing a soft robotics array for safe navigation in surgical process.

What it does

Each soft actuator can be actuated individually and reach a certain curvature. The five-segment array can be manipulated to point at a specific point on the plane. Potentiometers were used to control actuation of each segment.

How we built it

The process of developing such a system can be divided into three parts: fabrication and mechanical prototyping, electronics, and software. The overall system consists of the following components: a pump, 5 soft actuators, 3 common solenoids and 1 solenoid for each actuator, 1 potentiometer for each actuator, and 1 pressure transducer

The fabrication part involves designing and experimenting with a range of 3D-printed molds for fabricating the soft actuator. We came up with 2 different designs with varying sizes. One is rectangular and another is cylindrical and we experimented with different dimensions for both the designs. The next step is to fabricate the soft actuators using these molds and a soft material called ecoflex.
The hardware part involves a microcontroller and designing the circuits for mosfets, solenoids. potentiometers, pump and pressure transducers. The software part involves writing the software to control the mosfets which in turn control the solenoids connected to each of the actuators, take reading from potentiometers to control the movement of soft actuators, and read the pressure sensor to control the actuators without human interaction.

Challenges we ran into

For fabrication of soft actuators, because neither of us worked with solidworks before so we spent some time learning how to design molds in solidworks and make them into 3D-printing files. The 3D-printers were not always reliable thus we had a lot of failed molds. Because of the structure of the soft actuator, we had to divide the fabrication process into multiple steps: 3 steps for rectangular design and 4 steps for cylindrical design. Each step took at least 5 hours for the material to mix and cure. So it took a long time to fabricate actuators before testing if they work. Another big challenge is to make the actuator air-tight and able to stand pressure when fully actuated. Through experiment, we found out that the part where ecoflex and tube attach is the easiest to break under high pressure. We have tried multiple ways to enhance the attachment of ecoflex and tube, including using silicon epoxy, 3D-printed rings to make more exoflex wrap around the tube, heat shrink tubes, and hot glue. Some of the actuators eventually work, while some of them still failed the final testing. We were able to get enough working actuators for the final prototype; however this time consuming fabrication process prevented us from having more actuators to test for the final prototype to make it work better.

For hardware building, despite designing circuits for user input control and solenoids, we had challenge with the pressure transducer. The transducer we ordered can measure high pressure and output maximum 40 mV. So we built an instrumentation amplifier circuit to turn the small range output into MBED analog input. However, after studying it on our own as well as asking for help from our lab staff and TA, we found the transducer did not work as we expected it to according to the datasheet. We were able to get another type of pressure transducer from one of our classmate. It has an output range of 5 V and works well, but the pressure range is not big enough for our application. And it was not enough time for us too order new transducers. Thus we used the working pressure transducer to build the circuit and write the code for measuring pressure from actuators. However, because the pressure range is too small, we would not be able to test the feedback control until we get new working actuators.

Accomplishments that we're proud of

We were proud to learn the whole process of soft robotics fabrication and to be able to built air-tight actuators.

What we learned

• Design model in solidworks
• Use makerbot to 3D-print
• Fabricating soft robotics with ecoflex

What's next for Soft robotics arm

The next steps for this project involves getting working pressure transducers to test the feedback control code. More sensors and control feedback code can be added to the system to better control the overall movement of the actuator array. And for clinical purpose, a flashlight and a camera can be mounted to the end of the array to picture the surgery area. The diameter of the actuator should be decreased to under 12 mm, which is the current common limit for laparoscopic surgery. There are also problems need to be solve for soft robotics in general. For example, a more repeatable fabrication process needs to be developed to standardize performance of each actuator. Fatigue is another problem that soft robotics encounter when put into clinical application.

Built With

• mbed
• solidworks