In a recent article published in the journal Discover Applied Sciences, researchers proposed an approach that combines long-range wireless (LoRa) communication, simulation environments, digital twin concepts, and the Internet of Robotic Things (IoRT) to enable the remote operation of an agricultural robot with a digital twin/shadow. This method aims to assist the robot in navigating complex situations without high-end network infrastructure.
Study: Internet of robotic things with a local LoRa network for teleoperation of an agricultural mobile robot using a digital shadow. Image Credit: Suwin66/Shutterstock.com
Background
Agricultural robots are increasingly used to improve efficiency and reduce resource consumption in farming. However, autonomous navigation in unstructured and complex environments like orchards and fields requires human supervision and intervention. IoRT integrates robotics and wireless connectivity with virtual reality, cloud computing, and Internet of Things (IoT) platforms, enabling collaborative control and teleoperation of robots, along with real-time monitoring and data analysis.
IoRT has many potential applications in agriculture, such as precision farming, crop monitoring, harvesting, and spraying. However, it encounters challenges such as network availability, bandwidth limitations, latency issues, and the need for situational awareness and effective feedback for remote operators. To address these challenges, alternative solutions are required to ensure efficient and robust communication between robots and operators.
About the Research
In this paper, the authors explored the feasibility of utilizing a local LoRa network to facilitate uninterrupted communication between a simulated and a real robot in a berry field over distances of up to 3000 meters. The communication involved exchanging waypoint coordinates crucial for the robots' autonomous navigation from point A to point B.
LoRa, a low-power communication protocol operating in the 868 or 915 MHz frequency bands, offered a link budget of approximately 140 to 160 dB. The study also aimed to develop a "digital shadow" of the robot—a virtual representation that mirrors its actions and state—to enable effective monitoring and guidance.
The proposed system includes:
- A four-wheel drive and four-wheel steering mobile robot equipped with a GPS-based navigation controller, collision avoidance sensors, and a LoRa transceiver.
- A simulation environment with a digital twin/shadow of the robot and a pure-pursuit controller.
- A graphical user interface (GUI) for the remote operator.
- A set of LoRa transmitters and repeaters to ensure comprehensive wireless network coverage.
The system was built around a custom-designed circuit motherboard featuring dual microcontrollers, a real-time clock, onboard memory, and Controller Area Network (CANBUS) communication. The communication architecture was structured into four layers: the farm layer, backend layer, wrapper layer, and frontend layer, which together facilitated end-user access to live data and control commands.
The primary hypothesis tested was the robustness of the LoRa communication link between the operator and the robot, focusing on the control system's architecture, communication efficiency, and situational awareness. Several experiments were conducted to assess the performance of the proposed method under various field conditions and scenarios.
Research Findings
The outcomes revealed that the average packet loss was approximately 12 % at distances around 2300 meters, with variations depending on the environment. The digital shadow of the robot successfully followed a user-defined reference trajectory in real-time and generated a corresponding path in the geographic coordinate system.
Key waypoints of this path were transmitted as LoRa messages to the actual robot in the field, enabling it to update its path and navigate autonomously using its onboard navigation system. Additionally, the current position of the robot was sent as a LoRa message to the digital shadow, allowing for real-time updates to the simulation scene and the web-based map used for visualization.
Applications
This research enhances the safety and efficiency of robot navigation by providing a practical solution for remote intervention and supervision in challenging situations. It demonstrates the feasibility of creating a digital shadow of the robot, which supports path planning and tracking and provides real-time feedback and situational awareness to the remote operator.
The proposed method also has potential applications beyond agriculture, extending to other fields requiring LoRa communication and robot teleoperation, such as disaster response, search and rescue, and exploration.
Conclusion
In summary, the novel IoRT solution proved effective for the teleoperation of an agricultural robot using a digital shadow. It showed promising results in packet loss, path tracking, and situation awareness.
The researchers acknowledged limitations and challenges like stability, latency, and limited wireless communication range. They suggested that future work should consider using a 5G network to improve data transmission speed, navigation efficiency, and visual feedback. Additionally, they proposed conducting more extensive experiments and user studies to assess the method's usability and user satisfaction.
Journal Reference
Shamshiri, R.R., Navas, E., Dworak, V. et al. Internet of robotic things with a local LoRa network for teleoperation of an agricultural mobile robot using a digital shadow. Discov Appl Sci 6, 414 (2024). DOI: 10.1007/s42452-024-06106-7, https://link.springer.com/article/10.1007/s42452-024-06106-7
Disclaimer: The views expressed here are those of the author expressed in their private capacity and do not necessarily represent the views of AZoM.com Limited T/A AZoNetwork the owner and operator of this website. This disclaimer forms part of the Terms and conditions of use of this website.