Objective: Design and implementation of a Tactile Cyber-Physical System for real-time interaction between physical and virtual worlds for applications requiring ultra-reliable low latency communications
Time-sensitive networking (TSN) is the IEEE 802.1Q defined standard technology to provide deterministic messaging on standard Ethernet. It provides guarantees of delivery and minimized jitter using time scheduling for those real-time applications that require determinism.
- Time-Senstitive Networking (TSN) switch (Towards implementing IEEE 802.1 TSN):
- Time Synchronization (Towards implementing IEEE 802.1AS)
- Time Aware Shaper (Towards implementing IEEE 802.1Qbv)
- Packet duplication and Elimination (Towards implementing IEEE. 802.1CB)
- Tactile Cyber-Physical System:
- Real time control of a teleoperator robotic arm
- Using haptic devices to send kinematic data and receive haptic feedback
- Intelligence algorithms for performance improvement
- Edge intelligent switch ports on programmable switches using P4
Implemented Precision Time Protocol (PTP) on Linux devices. Tested using four Raspberry Pi devices – R1, R2, R3, and R4. R3 is the grand-master clock using the Best Master Clock Algorithm (BMCA). Remaining devices are in slave configuration. Using software timestamping a sub 100 μs synchronization error was obtained.
P4 based switch using Netronome Agilio SmartNIC:
We have setup a ethernet swtich using P4 programming language on the Netwronome Agilio SmartNIC. The match action table/rules contain the MAC addresses of the connected hosts. IP address is extracted from the packet and the packet is forwarded to a matching MAC address. The P4 program is also useful for obtaining and adding timestamp in a custom metadata header. Obtaining timestamps at the ports of the switch is a basic step for TSN.
Time Aware Shaper using LinuxTC on Netronome card:
Feedback to the human operator:
Control of UR3 robotic arm:
UR3 from Universal Robots is a high precision capable collaborative robot. It comes with a patented programming interface – PolyScope. But it has no real time control capability. A Python program using rospy library was developed for real time control of the robot using its external control feature. Some issues of vibration and jerky motion was fixed by developing few algorithms.
A demonstration of teleoperation of the UR3 robotic arm by a human operator using the Geomagic Touch haptic device:
Intelligent gripping of a tool:
Interfaced Robotiq 2F-85 two finger gripper to be controlled in real time. This allows us to remotely move position the robot to hold a tool and then pick it up by closing the gripper using a button input form the haptic device. The accuracy of gripping entirely depends on the human operator. To assist the operator, we provide haptic feedback from the gripping surfaces. This functionality is enabled using a force sensor array developed by us. The force data can be sent back to the haptic device as feedback. We use this data to intelligently grip a tool by using edge intelligence algorithms.
Ongoing and future work:
Soumya Kanta Rana