Authors
University/Institution
Brief Information about Fotokite
Fotokite is a Zurich CH, Syracuse NY, and Boulder CO based company creating cutting-edge tools that assist public safety teams with elevated situational awareness. They build fully autonomous, persistent, and reliable systems that provide invaluable overview information to firefighters and first responders to help them manage complex, safety-critical situations.
Introduction about Fotokite Sigma
Fotokite Sigma is a situational awareness system that consists of a remote-controlled drone and a tablet that shows live feed from the drone. Sigma can be used in many applications such as fire response, Search and Rescue operations, traffic management and much more.
Brief information about Stewart Platform:
As seen in the figure above; Stewart Platform consists of 6 arms that are prismatic joints powered by various types of motors. This design allows the moving platform to move in any direction and have a high payload at the same time.
A major design feature of parallel manipulators in general is that Inverse kinematics can be calculated easily; by having the angles of the motors as an input; you can calculate the position of the end effector (moving platform) as an output. Check out our blog for more information about Stewart Platforms:
https://acrome.net/post/what-are-hexapod-robots-used-for
Example Video: Using Stewart Platform in Fotokite's Sigma Drone Testing
Manufacturing Test Process of the Sigma Drones with the Stewart Platform
In Fotokite's manufacturing process, the Stewart Platform is used to calibrate all the Kite sensors by moving the system into different known poses. This calibration needs to be performed for each newly manufactured system. The Stewart Platform's control and feedback messages are integrated into the manufacturing software of Fotokite, which allows an automated calibration on the line. As the platform moves the Kite into various poses, the measurements of the sensors and the feedback values taken from the Stewart Platform are logged. This data is then used to calculate accurate, individual calibration parameters for each sensor of the Sigma drone.
A ready-to-use GUI is shipped and generally used for getting measurements of the sensors and the feedback values from the Stewart Platform directly. It is an out-of-the-box feature with no programming requirements.
Precision Calibration with Stewart Platform
Fotokite utilized the Stewart Platform to meet their specific manufacturing and calibration needs, as it allowed for precise movement and positioning in six dimensions (x, y, z, roll, pitch, and yaw). This measurement accuracy of the sensors is a crucial topic hence sensors in the Sigma drone, such as IMU and force sensors need to be calibrated at the factory. To achieve the highest degree of measurement precision, the Stewart Platform's top and bottom plates have to be modified to mount the drone as rigid as it could be, which is another essential feature. Additionally, the Stewart Platform's simple communication protocol with Python made it easy to integrate into their existing software.
You may check this link to learn more about Python support of ACROME products.
Fotokite initially used the first model of the Stewart Platform, but it proved not robust enough for their application, which required extensive and frequent use. They then switched to the Stewart Pro Platform, which offered the level of robustness for their needs. See the last section on how this has happened.
Future Plans for using Stewart Platforms in Fotokite's Processes
While Fotokite doesn't plan to develop new manufacturing processes involving the platform, they continue using the ACROME Stewart Pro Platforms for their calibration process’. It has been a nice experience till now, and we hope to see the Stewart Platforms serve Fotokite for manufacturing great products in the future as well.
Adapting ACROME's Stewart Platform to meet Fotokite's Requirements
Acrome started working with Fotokite back in 2018. We tried to achieve the desired positioning requirements and determine the factors affecting the precision even with the first order of the Fotokite company.
The sensors that are used in our Stewart Platforms provide analog output. Based on our measurements, we determined that these sensors don’t provide a linear enough output. We solved this challenge in 2 steps. Step#1, changing the system from Stewart to Stewart Pro models. Step#2, we developed a calibration method for a linearization process of each actuator. Then we created a "look-up table" which is used as the calibration data in our embedded firmware.
A specific requirement of the Fotokite was to distinguish and isolate the Stewart Platforms from other network devices that they are using. To achieve this, we changed the network IP addresses of each controller. We prepared special barcodes for the Stewart Platforms to ensure that the controllers match with their manufacturing line and this helps Fotokite to prevent the mismatch of controllers in different systems.”
