As 3D printing and robotics continue to collide and complement each other, new machines are being created. In soft robotics, we’re seeing the emergence of a class of machines that can deform or morph in response to their environments.
Rob Scharff, a PhD candidate at TU Delft, discusses more how these robots are able to change color via sensors and shape, moving further toward closed-loop control of soft robots, with further research published in ‘Color-based Proprioception of Soft Actuators Interacting with Objects.’ The researchers explain their goal to create soft robots that can interact with items that were previously unknown to them, using color-based sensors.
“Our results demonstrate that deformations of the actuator during interaction, including the sigmoid-like shapes, can be accurately reconstructed. The accurate shape sensing represents a significant step towards closed-loop control of soft robots in unstructured environments,” state the researchers.
Scharff explains soft robotics as the softer, more adaptive robot for today—as well as pointing out that their gripping devices have become much more attuned to handling items like fruits or vegetables, for example, making them useful for industries such as grocery stores or their warehouses—and in exercises where the items they are gripping could be unpredictable in size or fragility.
Why can’t we use traditional sensors in soft robots?
“Traditional robots typically have joints that revolve around a single axis,” says Scharff. “Therefore, a single encoder in each joint is enough to reconstruct the robot’s shape. In contrast, soft robots can bend, stretch and twist at the same time, making existing sensors unsuitable. Therefore, we need sensors that can capture a large variety of deformations. Besides that, the sensors need to be flexible so as not to impede the actuators movement.”
How do your sensors work?
“The bending actuators shown in this research consist of an air chamber with an inextensible layer at the bottom, and an extensible bellow-shape at the top. Inflating the air chamber will cause the bellows at the top to expand, while the bottom remains at the same length, creating a bending motion. We 3D print a color pattern inside these top bellows and observe these color patterns with color sensors at the inextensible bottom of the actuator. When the actuator is inflated, colors that were previously occluded from the sensors start to appear. We use this change in color and the changes in light intensity to predict the shape of the actuator.”
How do you derive the position from this color?
“We calibrate the sensors using a feedforward neural network. To train the network, we collected 1000 samples of sensor values with corresponding actuator shapes. The actuator shapes were represented by 6 markers on the inextensible layer that were tracked by a camera. The inputs of the network are the readings from 4 color sensors with 4 channels (red, green, blue, white) each.”
What is the accuracy of your new method?
“Our method is able to predict the position of each of the markers with an error that is typically between 0.025 and 0.075mm. Our method also performs well on load cases that were not present in the training data. Unlike existing sensors in soft robotics, we can measure the exact shape of how the gripper is bent around an object. Therefore, the work is a big step towards being able to accurately move and grasp objects with soft robots.”
How are the actuators fabricated?
“The actuators are fabricated as a single piece using PolyJet multi-material Additive Manufacturing. The bellows are fabricated using the flexible Agilus Black material, whereas the colored structure is composed of alternating segments of VeroCyan and VeroMagenta. The custom sensor PCB’s are embedded on small 3-D printed plugs that are plugged into the bottom of the actuator to create an airtight fit. Color sensors are very cheap and accessible which makes it easy to implement the technique in other soft robots as well.”
3D printing in multiple materials, as well as multiple colors, is becoming more common as technology and hardware catch up with concepts and user demands, along with the accompaniment of sensors, electronics, wearables, and much more. What do you think of this news? Let us know your thoughts; join the discussion of this and other 3D printing topics at 3DPrintBoard.com.[Source / Images: TU Delft]
You May Also Like
1960s Artwork Returns to Life With WASP’s Crane 3D Printing Technology
Once again, crane 3D printing company WASP captivates us with a new earthly design that blends art and culture with sustainable living. This time, the innovative Italian firm teamed up...
3D Printing News Briefs, July 11, 2021: Wohler’s Associates; Solvay, Ultimaker, and L’Oréal; America Makes & ODSA; BMW Group; Dartmouth College; BEAMIT & Elementum 3D; Covestro & Nexeo Plastics; Denizen
In today’s 3D Printing News Briefs, we’ll be telling you about the launch of an audio series and a competition, AM training and research efforts, materials, and more. Read on...
Intellegens Upgrades 3D Printing Deep Learning Software
As the first market research firm to publish a report on the rapidly evolving trend of automation in 3D printing, SmarTech Analysis noted how crucial new technologies like machine learning,...
MESO-BRAIN Uses Stem Cells & Nanoscale 3D Printing to Investigate Neural Networks
The MESO-BRAIN consortium is a collaborative research effort, led by the UK’s Aston University and funded by FET and the European Commission, that’s focused on developing 3D human neural networks...
View our broad assortment of in house and third party products.