Arvind Kumar Saxena
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Arvind Kumar Saxena has a total of 12 co-authors. Here are the top co-authors by number of shared publications:
- Parvindar Kaur Chhabda
Kalinga University
2 shared publications - Abdumutalliev Abdulakhad Abdusamad Ugli
Turan International University
1 shared publication Farida Karimova
Tashkent State University of Uzbek Language and Literature named after Alisher Navoi,Tashkent,Uzbekistan
1 shared publicationG. Prabhavathi
Godavari Global University,CSE(DS),Rajamahendravaram,Andhra Pradesh,533296
1 shared publicationG. Ravivarman
Karpagam Academy of Higher Education,Department of Electrical and Electronics Engineering,Coimbatore,641021
1 shared publicationHaider Mohammed Abbas
College of Technical Engineering, The Islamic University,Department of Computers Techniques Engineering,Najaf,Iraq
1 shared publication
Latest Publications
- 6 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 643-648
An Edge-Enabled Intelligent Real-Time Feedback Control System for Industrial Robotic Arms
Edge computing is changing the way industrial robots respond to their environments, making their movements faster and more accurate. An architecture that moves timecritical control tasks from the cloud to local servers so that managers no longer worry about signal delays or weak networks. By placing sensors, processors, and actuators in the same industrial cell, our framework delivers sub-millisecond actuation, greater fault tolerance, and steady performance whether payloads change suddenly or network traffic spikes. Motion plans are continuously refined through a closed loop that collects raw sensory data, analyses it on a lightweight edge node, and feeds adjustments back to the arm within the same computational cycle. A compact machine-learning engine running at the edge also flags anomalies and predicts upcoming disturbances, freeing operators to focus on strategic tasks. Tests on a six-degree-of-freedom industrial arm, controlled in a realistic production setting, show that feedback delay shrinks by up to 40 % and recovery from sensor faults takes almost half the time needed with standard Programmable Logic Controller. The solution upgrade brings the autonomy, speed, and safety needed by modern welding, assembly, and material-handling applications while meeting the strict timing deadlines set by safety standards. The proposed edge-enabled control system demonstrates significant quantitative gains over traditional architectures. It achieves a 40.7 % reduction in average latency $\mathbf{(} \mathbf{3. 2 ~ m s}$ vs. $\mathbf{5. 4 ~ m s}$ PLC), improves fault detection accuracy to 97.4 %, reduces recovery time by 49 %, and maintains low trajectory RMSE $(0.28-0.41 ~\text{mm})$ under varying load conditions.- 6 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 643-648
An Edge-Enabled Intelligent Real-Time Feedback Control System for Industrial Robotic Arms
Edge computing is changing the way industrial robots respond to their environments, making their movements faster and more accurate. An architecture that moves timecritical control tasks from the cloud to local servers so that managers no longer worry about signal delays or weak networks. By placing sensors, processors, and actuators in the same industrial cell, our framework delivers sub-millisecond actuation, greater fault tolerance, and steady performance whether payloads change suddenly or network traffic spikes. Motion plans are continuously refined through a closed loop that collects raw sensory data, analyses it on a lightweight edge node, and feeds adjustments back to the arm within the same computational cycle. A compact machine-learning engine running at the edge also flags anomalies and predicts upcoming disturbances, freeing operators to focus on strategic tasks. Tests on a six-degree-of-freedom industrial arm, controlled in a realistic production setting, show that feedback delay shrinks by up to 40 % and recovery from sensor faults takes almost half the time needed with standard Programmable Logic Controller. The solution upgrade brings the autonomy, speed, and safety needed by modern welding, assembly, and material-handling applications while meeting the strict timing deadlines set by safety standards. The proposed edge-enabled control system demonstrates significant quantitative gains over traditional architectures. It achieves a 40.7 % reduction in average latency $\mathbf{(} \mathbf{3. 2 ~ m s}$ vs. $\mathbf{5. 4 ~ m s}$ PLC), improves fault detection accuracy to 97.4 %, reduces recovery time by 49 %, and maintains low trajectory RMSE $(0.28-0.41 ~\text{mm})$ under varying load conditions.- 6 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1482-1488
Edge-Based Real-Time Analytics for Mobile Sports Performance Tracking
Mobile edge computing, IoT, and AI are innovative concepts in the fields of monitoring the mobile sports performance, with high hazards in comparison to the previous time regarding latency and data quality of the real-time analysis. However, all the existing solutions either are limited to the input of a single-sensor or limited to standalone metrics without the integration of different data streams that would generate the complete and functioning intelligence. Within this direction towards eliminating this dire gap, we introduce EdgeTrack, a new multi-modal analytics pipeline, entirely executed at the edge. The EdgeTrack is a combination of physiological (heart rate, SpO 2, etc.) and biomechanical (motion, acceleration, posture, etc.) sensor information and gives real time, contextually adaptive responses to athletes and coaches. It relies on a sensor fusion on-device, lightweight context-aware AI models, and resource allocation that can offer low-latency monitoring with high precision of various variables across various domains of sports. EdgeTrack, which can be used alongside the current edge System-on-Chips (SoCs) commercially, it does not have to connect to cloud computing infrastructure, it has a significantly faster response time without affecting the privacy of the user and wasting power. The platform of EdgeTrack has demonstrated itself in the practical environment of interactions with athletes of different sports to be more superior than the older system of cloud based and single modality. It realizes 60 percent, 12 percent more accurate and 35 percent lower energy consumption results. Although this is done individually, EdgeTrack will enable athletes to accept in-the-moment performance decisions and real-time data and coaches to have a better picture of the situation. It is a recent standard of intelligent sport tracking and the application is very extensive in team training, injury recovery and recreational exercise.
- 6 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1482-1488
Edge-Based Real-Time Analytics for Mobile Sports Performance Tracking
Mobile edge computing, IoT, and AI are innovative concepts in the fields of monitoring the mobile sports performance, with high hazards in comparison to the previous time regarding latency and data quality of the real-time analysis. However, all the existing solutions either are limited to the input of a single-sensor or limited to standalone metrics without the integration of different data streams that would generate the complete and functioning intelligence. Within this direction towards eliminating this dire gap, we introduce EdgeTrack, a new multi-modal analytics pipeline, entirely executed at the edge. The EdgeTrack is a combination of physiological (heart rate, SpO 2, etc.) and biomechanical (motion, acceleration, posture, etc.) sensor information and gives real time, contextually adaptive responses to athletes and coaches. It relies on a sensor fusion on-device, lightweight context-aware AI models, and resource allocation that can offer low-latency monitoring with high precision of various variables across various domains of sports. EdgeTrack, which can be used alongside the current edge System-on-Chips (SoCs) commercially, it does not have to connect to cloud computing infrastructure, it has a significantly faster response time without affecting the privacy of the user and wasting power. The platform of EdgeTrack has demonstrated itself in the practical environment of interactions with athletes of different sports to be more superior than the older system of cloud based and single modality. It realizes 60 percent, 12 percent more accurate and 35 percent lower energy consumption results. Although this is done individually, EdgeTrack will enable athletes to accept in-the-moment performance decisions and real-time data and coaches to have a better picture of the situation. It is a recent standard of intelligent sport tracking and the application is very extensive in team training, injury recovery and recreational exercise.