Venkatesan Kalaichelvi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Venkatesan Kalaichelvi has a total of 69 co-authors. Here are the top co-authors by number of shared publications:
- Ramanujam Karthikeyan
Department of Mechanical Engineering, Birla Institute of Technology and Science, Pilani, Dubai Campus, Dubai International Academic City, Dubai, United Arab Emirates
28 shared publications - Abhilasha Singh
Birla Institute of Technology and Science Pilani
8 shared publications Arvind Desikan
Department of Electrical and Electronics Engg., BITS Pilani, Dubai Campus, UAE
4 shared publicationsR.K. Ganesh Ram
BITS Pilani Dubai Campus
4 shared publicationsD. Sivakumar
Department of Electronics and Instrumentation Engineering, Annamalai University, Annamalai Nagar, Chidambaram, Tamilnadu, India
3 shared publicationsKarthikeyan Ramanujam
Department of Mechanical Engineering, Birla Institute of Technology and Science, Pilani, Dubai Campus, Dubai International Academic City, Dubai, United Arab Emirates
3 shared publications
Latest Publications
- Open Access
- 10 October 2025
- Published by Frontiers Media SA in Frontiers in Robotics and AI
- Vol. 12, 1612392
Adaptive emergency response and dynamic crowd navigation for mobile robot using deep reinforcement learning
Mobile robots have emerged as a reliable solution for dynamic navigation in real-world applications. Effective deployment in high-density crowds and emergency scenarios requires not only accurate path planning but also rapid adaptation to changing environments. However, autonomous navigation in such environments remains a significant challenge, particularly in time-sensitive applications such as emergency response. Existing path planning and reinforcement learning approaches often lack adaptability to uncertainties and time-varying obstacles, thereby making them less suitable for unstructured real-world scenarios. To address these limitations, a Deep Reinforcement Learning (DRL) framework for dynamic crowd navigation using three algorithms, Deep Deterministic Policy Gradient (DDPG), Twin Delayed Deep Deterministic Policy Gradient (TD3), and Deep Q-Network (DQN), is proposed. A context-aware state representation that combines Light Detection and Ranging (LiDAR)-based obstacle perception, goal orientation, and robot kinematics to enhance situational awareness is developed. The proposed framework is implemented in a ROS2 Gazebo simulation environment using the TurtleBot3 platform and tested in challenging scenarios to identify the most effective algorithm. Extensive simulation analysis demonstrates that TD3 outperforms the other approaches in terms of success rate, path efficiency, and collision avoidance. This study contributes a reproducible, constraint-aware DRL navigation architecture suitable for real-time, emergency-oriented mobile robot applications.
- 10 October 2025
- Published by Frontiers Media SA in Frontiers in Robotics and AI
- Vol. 12, 1612392
Open AccessAdaptive emergency response and dynamic crowd navigation for mobile robot using deep reinforcement learning
Mobile robots have emerged as a reliable solution for dynamic navigation in real-world applications. Effective deployment in high-density crowds and emergency scenarios requires not only accurate path planning but also rapid adaptation to changing environments. However, autonomous navigation in such environments remains a significant challenge, particularly in time-sensitive applications such as emergency response. Existing path planning and reinforcement learning approaches often lack adaptability to uncertainties and time-varying obstacles, thereby making them less suitable for unstructured real-world scenarios. To address these limitations, a Deep Reinforcement Learning (DRL) framework for dynamic crowd navigation using three algorithms, Deep Deterministic Policy Gradient (DDPG), Twin Delayed Deep Deterministic Policy Gradient (TD3), and Deep Q-Network (DQN), is proposed. A context-aware state representation that combines Light Detection and Ranging (LiDAR)-based obstacle perception, goal orientation, and robot kinematics to enhance situational awareness is developed. The proposed framework is implemented in a ROS2 Gazebo simulation environment using the TurtleBot3 platform and tested in challenging scenarios to identify the most effective algorithm. Extensive simulation analysis demonstrates that TD3 outperforms the other approaches in terms of success rate, path efficiency, and collision avoidance. This study contributes a reproducible, constraint-aware DRL navigation architecture suitable for real-time, emergency-oriented mobile robot applications.
- Open Access
- 18 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 127170-127182
Comparative Analysis of GPT-4 and LLaMA 3.2 Integration With Speech Processing Models for Enhancing Human–Robot Interaction and Motion Control in Real-World Applications
Human-Robot Interaction (HRI) in robots finds wide applications today such as personal assistant robots, autonomous vehicles, healthcare support robots, industrial robots and many more, which receive interpreted commands to perform functions ranging from home automation to real-time assembly line operations. This paper provides a comparative study of different Natural Language Processing (NLP) models that are created by combining advanced Large Language Models (LLMs) with speech processing technologies to create a more intuitive, adaptable, and accurate system for robots. By assessing performance metrics including accuracy, response time and robustness, this paper identifies the best generalized model for the real-world application in robotics. In fact, this framework enables natural language understanding and speech generation to be combined effectively, which can help robots respond quickly to spoken requests, even in dynamic environments.
- 18 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 127170-127182
Open AccessComparative Analysis of GPT-4 and LLaMA 3.2 Integration With Speech Processing Models for Enhancing Human–Robot Interaction and Motion Control in Real-World Applications
Human-Robot Interaction (HRI) in robots finds wide applications today such as personal assistant robots, autonomous vehicles, healthcare support robots, industrial robots and many more, which receive interpreted commands to perform functions ranging from home automation to real-time assembly line operations. This paper provides a comparative study of different Natural Language Processing (NLP) models that are created by combining advanced Large Language Models (LLMs) with speech processing technologies to create a more intuitive, adaptable, and accurate system for robots. By assessing performance metrics including accuracy, response time and robustness, this paper identifies the best generalized model for the real-world application in robotics. In fact, this framework enables natural language understanding and speech generation to be combined effectively, which can help robots respond quickly to spoken requests, even in dynamic environments.
- Open Access
- 1 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 176287-176326
A Comprehensive Survey of Path Planning Algorithms for Autonomous Systems and Mobile Robots: Traditional and Modern Approaches
- 1 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 176287-176326
Open AccessA Comprehensive Survey of Path Planning Algorithms for Autonomous Systems and Mobile Robots: Traditional and Modern Approaches
- 24 April 2024
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 84 (10) , 7377-7400
Performance evaluation of vision based path planning for dynamic real-time scenarios of mobile robot
- 24 April 2024
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 84 (10) , 7377-7400
Performance evaluation of vision based path planning for dynamic real-time scenarios of mobile robot
- 11 December 2023
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 83 (18) , 56229-56246
Machine learning-based multi-sensor fusion for warehouse robot in GPS-denied environment
- 11 December 2023
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 83 (18) , 56229-56246
Machine learning-based multi-sensor fusion for warehouse robot in GPS-denied environment
- 7 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 179-184
Crop Disease Detection and Classification using Transfer Learning and Deep Convolutional Neural Networks
Crop diseases pose a significant threat to global food security, affecting the livelihoods of millions of smallholder farmers. Identifying these diseases is crucial yet challenging due to lack of proper infrastructure. Leveraging the widespread adoption of smartphones and with the advancements in computer vision, this paper proposes a solution through smartphone-assisted disease diagnosis. Utilizing a publicly available dataset comprising of 54,306 images which include both healthy and diseased plants, we introduce a deep convolutional neural network which can identify 14 crop species and 38 diseases. Our final solution achieves an accuracy of 99.2% in training and 95.62% in testing, hence demonstrating the feasibility of the proposed method. Further improvement is possible with the help of a more diverse dataset. Overall, the paper outlines a clear pathway for global crop disease detection through smartphones.
- 7 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 179-184
Crop Disease Detection and Classification using Transfer Learning and Deep Convolutional Neural Networks
Crop diseases pose a significant threat to global food security, affecting the livelihoods of millions of smallholder farmers. Identifying these diseases is crucial yet challenging due to lack of proper infrastructure. Leveraging the widespread adoption of smartphones and with the advancements in computer vision, this paper proposes a solution through smartphone-assisted disease diagnosis. Utilizing a publicly available dataset comprising of 54,306 images which include both healthy and diseased plants, we introduce a deep convolutional neural network which can identify 14 crop species and 38 diseases. Our final solution achieves an accuracy of 99.2% in training and 95.62% in testing, hence demonstrating the feasibility of the proposed method. Further improvement is possible with the help of a more diverse dataset. Overall, the paper outlines a clear pathway for global crop disease detection through smartphones.
- Open Access
- 1 April 2023
- Published by Walter de Gruyter GmbH in Transport and Telecommunication Journal
- Vol. 24 (2) , 110-119
Ego Vehicle Lane Detection and Key Point Determination Using Deep Convolutional Neural Networks and Inverse Projection Mapping
Sciendo provides publishing services and solutions to academic and professional organizations and individual authors. We publish journals, books, conference proceedings and a variety of other publications.
- 1 April 2023
- Published by Walter de Gruyter GmbH in Transport and Telecommunication Journal
- Vol. 24 (2) , 110-119
Open AccessEgo Vehicle Lane Detection and Key Point Determination Using Deep Convolutional Neural Networks and Inverse Projection Mapping
Sciendo provides publishing services and solutions to academic and professional organizations and individual authors. We publish journals, books, conference proceedings and a variety of other publications.
- Open Access
- 1 April 2023
- Published by Institute of Advanced Engineering and Science in International Journal of Electrical and Computer Engineering (IJECE)
- Vol. 13 (2) , 1522-1532
Mathematical modeling and kinematic analysis of 5 degrees of freedom serial link manipulator for online real-time pick and place applications
Modeling and kinematic analysis are crucial jobs in robotics that entail identifying the position of the robot’s joints in order to accomplish particular tasks. This article uses an algebraic approach to model the kinematics of a serial link, 5 degrees of freedom (DOF) manipulator. The analytical method is compared to an optimization strategy known as sequential least squares programming (SLSQP). Using an Intel RealSense 3D camera, the colored object is picked up and placed using vision-based technology, and the pixel location of the object is translated into robot coordinates. The LOBOT LX15D serial bus servo controller was used to transmit these coordinates to the robotic arm. Python3 programming language was used throughout the entire analysis. The findings demonstrated that both analytical and optimized inverse kinematic solutions correctly identified colored objects and positioned them in their appropriate goal points.
- 1 April 2023
- Published by Institute of Advanced Engineering and Science in International Journal of Electrical and Computer Engineering (IJECE)
- Vol. 13 (2) , 1522-1532
Open AccessMathematical modeling and kinematic analysis of 5 degrees of freedom serial link manipulator for online real-time pick and place applications
Modeling and kinematic analysis are crucial jobs in robotics that entail identifying the position of the robot’s joints in order to accomplish particular tasks. This article uses an algebraic approach to model the kinematics of a serial link, 5 degrees of freedom (DOF) manipulator. The analytical method is compared to an optimization strategy known as sequential least squares programming (SLSQP). Using an Intel RealSense 3D camera, the colored object is picked up and placed using vision-based technology, and the pixel location of the object is translated into robot coordinates. The LOBOT LX15D serial bus servo controller was used to transmit these coordinates to the robotic arm. Python3 programming language was used throughout the entire analysis. The findings demonstrated that both analytical and optimized inverse kinematic solutions correctly identified colored objects and positioned them in their appropriate goal points.
- 6 January 2023
- Published by Taylor & Francis in International Journal of Computer Integrated Manufacturing
- Vol. 36 (7) , 1083-1108
Performance analysis of object detection algorithms for robotic welding applications in planar environment
Weld seam detection and tracking using intelligent techniques is a common task in robotic welding. In this study, vision-based robotic welding is implemented using a 2D Camera to reduce the experimental cost in a planar environment. Firstly, different multi-class object recognition algorithms namely Faster Region-Based Convolutional Neural Network (Faster R-CNN), Single Shot Detector (SSD) and You Look Only Once (YOLOv3) are trained with 1530 manually labeled images. To recognize the weld seam under different brightness levels addressing arc light and splash, Contrast-Limited Adaptive Histogram Equalization (CLAHE) method has been used. Further, an extensive experimental study is conducted to find the optimal hyperparameters among three deep learning models. The performance metrics like Accuracy, Root Mean Square Error, Precision, Recall, Mean Average Precision and F1-score are calculated for different test cases. Finally, homogeneous coordinate transformation is implemented to obtain robot coordinates from weld seam pixel coordinates. The analysis is also extended to detect the bad welding conditions for improper shapes and track the weld seam for good welding shapes using TAL BRABO manipulator. It was found that the weld shapes were accurately detected and tracked precisely with 99.9% accuracy using YOLOv3 than Faster R-CNN and SSD.
- 6 January 2023
- Published by Taylor & Francis in International Journal of Computer Integrated Manufacturing
- Vol. 36 (7) , 1083-1108
Performance analysis of object detection algorithms for robotic welding applications in planar environment
Weld seam detection and tracking using intelligent techniques is a common task in robotic welding. In this study, vision-based robotic welding is implemented using a 2D Camera to reduce the experimental cost in a planar environment. Firstly, different multi-class object recognition algorithms namely Faster Region-Based Convolutional Neural Network (Faster R-CNN), Single Shot Detector (SSD) and You Look Only Once (YOLOv3) are trained with 1530 manually labeled images. To recognize the weld seam under different brightness levels addressing arc light and splash, Contrast-Limited Adaptive Histogram Equalization (CLAHE) method has been used. Further, an extensive experimental study is conducted to find the optimal hyperparameters among three deep learning models. The performance metrics like Accuracy, Root Mean Square Error, Precision, Recall, Mean Average Precision and F1-score are calculated for different test cases. Finally, homogeneous coordinate transformation is implemented to obtain robot coordinates from weld seam pixel coordinates. The analysis is also extended to detect the bad welding conditions for improper shapes and track the weld seam for good welding shapes using TAL BRABO manipulator. It was found that the weld shapes were accurately detected and tracked precisely with 99.9% accuracy using YOLOv3 than Faster R-CNN and SSD.
- Open Access
- 7 November 2022
- Published by MDPI AG in Electronics
- Vol. 11 (21) , 3636
A Vision-Based Bio-Inspired Reinforcement Learning Algorithms for Manipulator Obstacle Avoidance
Path planning for robotic manipulators has proven to be a challenging issue in industrial applications. Despite providing precise waypoints, the traditional path planning algorithm requires a predefined map and is ineffective in complex, unknown environments. Reinforcement learning techniques can be used in cases where there is a no environmental map. For vision-based path planning and obstacle avoidance in assembly line operations, this study introduces various Reinforcement Learning (RL) algorithms based on discrete state-action space, such as Q-Learning, Deep Q Network (DQN), State-Action-Reward- State-Action (SARSA), and Double Deep Q Network (DDQN). By positioning the camera in an eye-to-hand position, this work used color-based segmentation to identify the locations of obstacles, start, and goal points. The homogeneous transformation technique was used to further convert the pixel values into robot coordinates. Furthermore, by adjusting the number of episodes, steps per episode, learning rate, and discount factor, a performance study of several RL algorithms was carried out. To further tune the training hyperparameters, genetic algorithms (GA) and particle swarm optimization (PSO) were employed. The length of the path travelled, the average reward, the average number of steps, and the time required to reach the objective point were all measured and compared for each of the test cases. Finally, the suggested methodology was evaluated using a live camera that recorded the robot workspace in real-time. The ideal path was then drawn using a TAL BRABO 5 DOF manipulator. It was concluded that waypoints obtained via Double DQN showed an improved performance and were able to avoid the obstacles and reach the goal point smoothly and efficiently.
- 7 November 2022
- Published by MDPI AG in Electronics
- Vol. 11 (21) , 3636
Open AccessA Vision-Based Bio-Inspired Reinforcement Learning Algorithms for Manipulator Obstacle Avoidance
Path planning for robotic manipulators has proven to be a challenging issue in industrial applications. Despite providing precise waypoints, the traditional path planning algorithm requires a predefined map and is ineffective in complex, unknown environments. Reinforcement learning techniques can be used in cases where there is a no environmental map. For vision-based path planning and obstacle avoidance in assembly line operations, this study introduces various Reinforcement Learning (RL) algorithms based on discrete state-action space, such as Q-Learning, Deep Q Network (DQN), State-Action-Reward- State-Action (SARSA), and Double Deep Q Network (DDQN). By positioning the camera in an eye-to-hand position, this work used color-based segmentation to identify the locations of obstacles, start, and goal points. The homogeneous transformation technique was used to further convert the pixel values into robot coordinates. Furthermore, by adjusting the number of episodes, steps per episode, learning rate, and discount factor, a performance study of several RL algorithms was carried out. To further tune the training hyperparameters, genetic algorithms (GA) and particle swarm optimization (PSO) were employed. The length of the path travelled, the average reward, the average number of steps, and the time required to reach the objective point were all measured and compared for each of the test cases. Finally, the suggested methodology was evaluated using a live camera that recorded the robot workspace in real-time. The ideal path was then drawn using a TAL BRABO 5 DOF manipulator. It was concluded that waypoints obtained via Double DQN showed an improved performance and were able to avoid the obstacles and reach the goal point smoothly and efficiently.