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


Gidugu Lakshmi Srinivas has a total of 18 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Real-Time Detection and Localization of Force on a Capacitive Elastomeric Sensor Array Using Image Processing and Machine Learning

    by Peter Werner Egger, Gidugu Lakshmi Srinivas, Mathias Brandstötter

    Soft and flexible capacitive tactile sensors are vital in prosthetics, wearable health monitoring, and soft robotics applications. However, achieving accurate real-time force detection and spatial localization remains a significant challenge, especially in dynamic, non-rigid environments like prosthetic liners. This study presents a real-time force point detection and tracking system using a custom-fabricated soft elastomeric capacitive sensor array in conjunction with image processing and machine learning techniques. The system integrates Otsu’s thresholding, Connected Component Labeling, and a tailored cluster-tracking algorithm for anomaly detection, enabling real-time localization within 1 ms. A 6×6 Dragon Skin-based sensor array was fabricated, embedded with copper yarn electrodes, and evaluated using a UR3e robotic arm and a Schunk force-torque sensor to generate controlled stimuli. The fabricated tactile sensor measures the applied force from 1 to 3 N. Sensor output was captured via a MUCA breakout board and Arduino Nano 33 IoT, transmitting the Ratio of Mutual Capacitance data for further analysis. A Python-based processing pipeline filters and visualizes the data with real-time clustering and adaptive thresholding. Machine learning models such as linear regression, Support Vector Machine, decision tree, and Gaussian Process Regression were evaluated to correlate force with capacitance values. Decision Tree Regression achieved the highest performance (R2=0.9996, RMSE=0.0446), providing an effective correlation factor of 51.76 for force estimation. The system offers robust performance in complex interactions and a scalable solution for soft robotics and prosthetic force mapping, supporting health monitoring, safe automation, and medical diagnostics.


  • Open Access

    Five-Axis Printing of Continuous Fibers on the Mold

    by Paweł Michalec, Marius Laux, Gidugu Lakshmi Srinivas, Robert Weidner, Mathias Brandstötter

    This paper explores a five-axis printing method designed to improve the fabrication of continuous fiber-reinforced thermoplastic composites (CFRTPCs), essential for producing lightweight, complex structures in advanced manufacturing. Traditional CFRTPC placement techniques often face challenges with precision, scalability, and optimal fiber orientation, especially in customized, small-scale applications. The proposed five-axis printing technique overcomes these issues by enabling precise fiber orientation and the production of robust spatial structures using 3D-printed molds compatible with CFRTPCs. Validation through three-point bending and surface quality tests revealed that five-axis printed cylindrical-lattice samples, with fibers oriented at 45°, exhibited superior mechanical properties and surface quality. The five-axis printed samples achieved a load-to-weight ratio 27% higher than traditional samples and maintained their shape even under significant deformation. Surface quality improved significantly, with roughness values reduced from 37.63 µm to approximately 12 µm. This method advances CFRTPC applications in industries requiring complex, lightweight components.


  • Development and Characterization of Tactile Capacitive Sensor for Soft Robotics Application

    by Gidugu Lakshmi Srinivas

    This paper details developing and characterizing a soft capacitive sensor patch designed for force measurement in soft robotics and orthosis applications. The 6x6 matrix sensor was fabricated using silicone, selected for its viscoelasticity and strain resistance, making it well-suited for human-interactive applications. The sensing mechanism is based on capacitance changes when force is applied and measured using a MUCA board. The system also incorporates an SHT30 sensor to monitor environmental conditions like temperature and humidity, ensuring the sensor’s performance under varying conditions. Data from the capacitive sensor is processed via an Arduino Nano 33 IoT, with all measurements wirelessly transmitted and displayed in real-time using a custom dashboard on Arduino Cloud. The sensor can detect forces in the range of 1 to 4 N. Machine learning techniques, specifically linear regression, were used to characterize the sensor, yielding a highly accurate R value of 0.9932. Capacitance and environmental data are automatically saved and processed using Python scripts, allowing for efficient data analysis. This capacitive sensor shows great promise in applications requiring precise force measurement and enhanced human-machine interaction. Potential future uses include wearable orthosis devices to improve human comfort and soft robotic systems designed to mimic natural movements. The ability to integrate this sensor into wearable technology highlights its potential in fields such as medical rehabilitation and prosthetics, offering improved interaction and adaptability in human-centered environments.

  • Improving Soft Capacitive Tactile Sensors: Scalable Manufacturing, Reduced Crosstalk Design, and Machine Learning

    by Gidugu Lakshmi Srinivas, Sherjeel Khan

    Soft and flexible capacitive tactile sensors are an important tool to; accurately measure tactile forces in wearable health monitors and enable soft grasping in robots. Precise force measurements in real-time pave the way for enhanced human-machine interaction, improved automation safety, and novel medical diagnostics approaches. In this work, a parallel plate soft capacitive sensor array was fabricated using low-cost and scalable additive manufacturing-based methods (screen printing and spin coating) with EMI shielding. Experimental setups were developed to evaluate the applied force on the soft, flexible sensor based on the ratio of mutual capacitance (RMC). The preciseness of the sensor array was tested with a normal force testing setup using Zwick Roell’s benchtop testing machine. Furthermore, a Robot was used to apply the normal force on each sensor ranging between 1–15 N, and its values were recorded using a force and torque sensor. IoT-based electronics were used to measure the RMC from five contact points simultaneously and a Python script saved data over time. The values of RMC were mapped to their corresponding force values using linear regression algorithms. This work demonstrates the suitability of the proposed sensor design for various application areas such as soft robotics and prosthetics for complex force measurements in human-machine interaction scenarios.


  • Open Access

    Versatile 3D-printed fin-ray effect soft robotic fingers: lightweight optimization and performance analysis

    by Gidugu Lakshmi Srinivas, Arshad Javed, Lisa Marie Faller

    Fin ray soft robotic fingers are inspired by the structure and movement of fish fins, enabling flexible and adaptive grasping capabilities. Addressing the challenges of resource efficiency in terms of reduced energy consumption and material expense, this work focuses on further optimizing inherently low-energy fin-ray fingers towards lightweight design. Soft grippers are used frequently in dynamically changing environments and have become inevitable in handling tasks for delicate objects. However, these grippers generally show limited performance and payload-carrying capacity in high-force application scenarios. To address these limitations, topology optimization technique is used here to obtain both gripping capabilities and high factor of safety (FOS) of fingers. The performance of various structures of fin-ray and optimized fingers are analyzed: rectangular, trapezoidal, straight struts, and inclined struts for angles + 45°, − 45°. The topologically optimized structure has 15.2% less mass compared to considered fin-ray finger’s average mass. The deflection coefficient (Cd) is calculated to select the best structure of the fingers based on grasping scenario, and its value should be minimum. The straight strut finger with thickness of t = 2 mm shows best wrapping capabilities compared to all fingers with Cd = 0.1574. The topologically optimized finger’ Cd = 0.1896 at volume fraction of 0.1. Even though the Cd is slightly higher, its FOS is 1.71 times higher. An experimental setup is developed to validate the simulation results with the help of a UR3e robotic arm and an AXIA80 force sensor. The grasping demonstration of soft robotic gripper is performed on various objects: coffee cup and wooden block.


  • Multi-axis Additive Manufacturing: Development of Slicer and Toolpath for 2.5D/3D/5D Printing

    by Gidugu Lakshmi Srinivas, Marius Laux, Vishnu Parameswaran Nair, Mathias Brandstötter


  • Design and Development of Fin-Ray Finger using Topology Optimization with Multiple Load Cases

    by Gidugu Lakshmi Srinivas, Alshawabkeh Mohammad, Mathias Brandstötter

    This research focuses on developing a soft robotic gripper using topology optimization method to address the need for soft and flexible grippers capable of delicately grasping complex objects with high conformability. The gripper design draws inspiration from the triangular shape of fin-ray fingers with rectangular crossbeams, mimicking the physiology of fish fins traditionally employed for soft grasping applications. However, existing design methods face limitations in high-load applications due to structural stiffness constraints. To overcome this limitation, this study proposes a mathematical reconstruction of the design framework using the topology optimization method. The objective function is formulated to maximize deflection, ensuring optimal wrapping capabilities of the finger. Constraints include the force vector, global stiffness, and volume fraction (v f ). The study incorporates multiple load cases to simulate the dynamic forces experienced by the robotic fingers during object manipulation in a changing environment. Using mathematical gradients, various fingers are developed for different volume fractions and rectangular crossbeams of fin-ray effect. The MATLAB code can adopt any mesh size and calculate the performance values of the finger. The optimized fingers exhibit maximum deflection with superior wrapping capabilities. The computational time for developed topology optimization algorithm is less than 100 sec. The best fingers are selected and 3D printed using TPU 95A material for validating simulation results and demonstrating the grasping tasks using Ultimaker S3 printer and Universal robot (UR3e). The simulation and experimental results exhibit a strong agreement, affirming the effectiveness of the proposed design approach.


  • Supportless 5-Axis 3D-Printing and Conformal Slicing: A Simulation-based Approach

    by Lakshmi Srinivas G, Michalec Pawel, Laux Marius, Lisa Marie Faller

    In the conventional three-dimensional (3D) printing, the parts are generated layer-by-layer along the z-axis by depositing the material. This approach simplifies the fabrication process, kinematics, and mechanical design of the 3D printers. However, the printed parts have non-uniform structural strength, planar visible layers, more travelling time, creates seam and support structures, etc. At the same time, this approach leads to the additional material and processing time. To overcome these issues, researcher focuses on the development of multi axis computer numerical control (CNC) and more than 6 degree of freedom (DOF) industrial manipulators as 3D-printers. However, such multi axis 3D-printing is still inaccessible for industrial use because of non-availability of fully developed slicing software. To make multi axis 3Dprinting more accessible, in this paper, we present a novel 5-axis conformal slicing algorithm for 3D-printing. The 5-axismaker machine is used to demonstrate 3D-printing of non-linear objects using iso-curves. Initially, the forward kinematics of the machine are computed using Denavit–Hartenberg (DH) parameter method, and motor variables are computed using inverse kinematics approach. Later, interface for conformal slicer is developed with the help of Rhino3D and Grasshopper software to generate the G-Code. The developed slicer allows the full control of printing path and parameters. Finally, the tubes (loft) are designed and printed with 3-axis and 5-axis methods to demonstrate the feasibility of algorithms. The ANSYS simulations are conducted for static study to check the structural strength of the parts. The simulations are validated by conducting experiments for compression test using Zwick Roell Z020. The results from both simulations and experiments are in good agreement, with a deviation of only 3.41%.


  • Parametric analysis of topologically optimized mechanical member considering dynamic loading

    by G Lakshmi Srinivas, Arshad Javed

    The optimal topology and its performance in dynamic loading situations result in discontinue function corresponding to the input factors such as volume fraction, thickness, material property, and loading conditions. In a realist scenario, the performance prediction becomes erroneous and challenging for the components under dynamic loading conditions with uncertainties. The conventional closed-form deterministic approaches are complicated for these problems. Here, a method is presented to establish the relative influence and function relationship of the input factors with the performance values, including controllable and non-controllable uncertainties. The design of experiment approach is used to apply full factorial design with Taguchi’s orthogonal array; performances of the optimal topologies are considered responses. The non-uniform topology generation method is applied based on the deflection threshold value to generate topologies for dynamic conditions. A dynamic model of the manipulator-link is developed to apply boundary conditions and provide performance values: compliance, deflection, Stress, and energy consumption values. Statistical techniques such as the analysis-of-mean (ANOM), analysis-of-variance (ANOVA), signal-to-noise-ratio (SNR), and mean performance values are employed to observe the significance of input factors and generate equivalent preformation relation. From ANOM and ANOVA, all input parameters show mutual interaction; force is observed as the most significant factor. From SNR values, experimental combination number 9,9,6,1 is observed as the most robust for compliance (21.13), deflection (43.93), Stress (−16.64), and energy consumption (12.05). Similarly, at the same combinations, the mean performance values are minimum and coefficient of determination (R2) percentages of the model are 94.64%, 96.93%, 73.69%, and 95.14%.