In less than 30 minutes of installation and calibration, we managed to integrate RealSense Depth Camera D435 and Adheso gripper with Meca500.

The small size and lightweight factors allow the camera to be attached on the arm. The Depth Camera has a range of 0.2m to 10m varying according to the light conditions. It has a depth diagonal Field of View over 90°

Because of its intelligent AI based vision system from Solomon Accupick, it doesn't require any training before running the program. Not only it can give robot "eyes" to see, but also provide them with a "brain" to understand with human-like cognition.

Meca500 is always a slave and never a master. Therefore, it can connect to any accessories/ peripherals as long as it is compatible with the PLC.

 

 

When the photoelectric sensor on the conveyor belt detected an item, the stepper motor will stop. This allows Meca500 to collect the packet of biscuits and store it in a container.

In this video, I had demonstrated 2 different types of ways to release the object (rotating and tilting). As the holder is flexible, adjustments need to be done to prevent collisions from happening.

After the container is full, it will proceed to the next step.

 

 

The angle decreases gradually when Meca500 tightens the screw onto the board using MEGP 25E electric gripper. Besides this application, there are also 2 more options available.

The first option, using the sixth joint of Meca500 to rotate ±100 revolution. This is not designed for this purpose, especially if the torque needs to be very precise.

The second option is to use a robotic screwdriver. By attaching it to the flange of Meca500, it can use its magnetized bit to pick up the screw. It is easy to use and we recommend this solution to our customers.

 

 

Creating a Graphical User Interface (GUI) using Python to control Meca500 to grasp the tiles from its original position to its desination.

Setting a new reference frame according to the robot's current location of the Tool Center Point (TCP) allow us to locate all the desired location with just one known position.

In this demostration, the colors of the tiles are arranged by Meca500 according to the arrangement stated in the GUI. The number of tiles left on the platform are also reflected in the interface to prevent the user from over selecting it. The GUI itself is programmed in a way that it will prompt the user to select other colors after calculating.

 

 

The embedded vision system in the Sawyer allows the arm to adjust to changes mid-task by identifying the displacement of the landmarks attached to the Picking and Placement Trays. As long as the landmark remains within the frame of the Sawyer's vision system, it will be able to accomodate for any slight changes in its given task.

The Sawyer's Pattern Recognition function also allows the cobot to carry out the pick-and-place task without the need for any hard-coding. Here we can see that the first position of the arm intiates the Pattern Algorithm, after adjusting to the changes, the arm indexes and shifts to the second node for each of the picking and placement process. Check it out "here"

In this video we demonstrate how the Meca500 can be controlled by a Raspberry Pi while monitoring its task - to do an accuracy test using a M7 bolt, remotely from another computer. The circuit consists of a Time of Flight (ToF) sensor (TeraRanger Evo 3m) from Terabee, the Meca500 robotic arm, and a simple LED traffic light tree

The remote access of the Raspberry Pi essentially gives me wireless control over the Meca500, bypassing its conventional need of the ethernet TCP, for communication with the monitoring device itself. The Pi can be controlled via the SSH or VNC servers. Then with the python script available for use with the Terabee sensor (open source on Github) [1], I developed a new algorithm for use with the entire circuit. Using Python's matplotlib function, I enabled a plot of 2 points/s with the data fed by the ToF sensor and displayed it on a separate monitor that is controlling the Meca500 via the Raspberry Pi, remotely.

An intrusion detected by the sensor (distance between hand and sensor falls below a cetain range), will send a signal to the light tree LED which indicates a red light, at the same time commanding Meca500 to stop. Removing the obstacle will allow the robotic arm to continue its task indicated by a yellow light.

References

[1] Terabee Sample Codes. (n.d.). Retrieved from GitHub: https://github.com/Terabee/sample_codes/blob/master/Python/Evo_64px_visualization_py3.py

 

 

 

ArtiMinds - a simulation software that supports multiple robots with various accessories.

With this simulation software, users are able to simulate motions of ABB, Denso, Fanuc, KUKA, Mecademic, UR for offline programming.

Other than the robots, ArtiMinds also integrate it with various brands of end-effector, vision, force/torque sensor to plan the motion ahead.

In this recording, the objects are assembled in sequence and random order respectively. As the motions are morphed into the hierarchy for simplification purposes, it is easier to change the order of templates.

 

 

Simulating carving motion using Artiminds software allows us to import any STEP/ CAD file to create a continuous path with just a few clicks on the setting. It is easy to define and change the tool-paths based on the uploaded file.

Artiminds also include features like collision-free motion planning, reachability assistants, collision checks, and calibration tools. Setting up new templates of mathematics precision with the edges can be done without teaching.

Sawyer - a collaborative robot that prioritizes safety over other factors that allow it to work in a dangerous or monotonous environment for humans.

The in-built Cognex camera at the arm allows the robot to locate/ inspect items without extra components. With this feature, Sawyer is able to grasp the object within the visual frame. This provides a dynamic reorientation and easy redeployment of the robot.

Afterwhich, Sawyer will place the object to its designated position whereby force sensing is applied to its surrounding. The cobot will repeat the force sensing loop until it finally finds the correct spot to place. Sensitive torque sensors embedded into every joint allows constant force control where delicate part insertion is critical, or to use force feedback in tasks where verification of properly seated parts is needed. Sawyer controls the amount of force it applies to different directions, the same way people do when performing tasks, eliminating the need for additional sensors or hardware.

 

 

Millions of smart phones are being produced by various manufacturers each year. Especially after the market starts to recover from the initial impact of the pandemic, the production of smart phones will increase in every major region.

Through the assistance of ADHESO gripper, Meca500 can help to assemble smart phone without using different end effector to pick various materials resulting in better productivity.

To optimize the efficiency of the workflow, the speed of Meca500 can increase from 20% to 70%.

For more information, please refer to our article "ADHESO Gripper that adapts according to the workpiece"

 

 

 

In this demonstration, Meca500 is transferring the wafer from one location to the FOUP for the further process to be done in the cleanroom.

In order to separate the wafer from the gripper, Meca500 has to tilt its fifth joint to simulate this motion. As the ADHESO gripper is particle-free, it does not leave any residue on the surface.

For more information, please refer to our article "ADHESO Gripper that adapts according to the workpiece"

 

 

 

In this demonstration, Meca500 will make use of it's 5μm position repeatability to place the IC chip on the breadboard.

With the help of the ADHESO gripper, Meca500 can easily collect the IC chip from the rack. This can help to check if the pins are straight before sending the chip out of the manufacturing company.

For more information, please refer to our article "ADHESO Gripper that adapts according to the workpiece"

 

 

 

The ADHESO gripper makes use of Van der Waals technology to stack the discs without leaving any blemish on the surface. In this demonstration, Meca500 is transferring discs from CD case to CD holder with disc protector in between.

For more information, please refer to our article "ADHESO Gripper that adapts according to the workpiece"

 

 

 

When a person approaches the collection point at a close distance, a signal is sent from Arduino to the Sawyer to dispense a waterbottle. The tower light will prompt the individual when can he take the waterbottle. Red means busy, Yellow means idling, Green means to take the bottle.

The graph shows the distance measured by the Terabee sensor

For more information, please refer to our article "How Internet of Things Can Improve Our System"

 

Integrating Meca500 with Solomon's 3D Vision System allows objects with different orientations to be placed back into preset position.

Executing the flowchart can be done by the drag-and-drop method without programming in the robot's language. The flowcharts can cater to any processor by selecting the robot brand.

Besides integrated vision, Meca500 can be incorporated with a suction end effector. We designed the plate adapter to fit Vuototecnica vacuum cup holder to the robot's last joint, allowing the object to be picked up from one location to another using the suction.

After Meca500 had placed the cubes into position, Sawyer will be activated to pick and place it to the designated locations