Learning Kits & Research Tools
Hands-on systems bridging theoretical knowledge with practical applications, supporting next-generation engineering education across India.
Machine Learning Kit
A compact, sensor-rich system generating real-world data from a DC motor — the complete ecosystem bridging embedded hardware with intelligent no-code software.
Data Collection & Preprocessing
Real-time gathering from multiple sensors; cleaning and preprocessing workflows.
Exploratory Data Analysis
Visualisation techniques to develop critical data interpretation skills.
Feature Engineering
Identify, create, and select relevant features from raw sensor data.
Model Development & Evaluation
Apply ML algorithms from basic regression to advanced classification.
Real-world Problem Solving
Tackle realistic scenarios with actual sensor data; bridge theory to practice.
Algorithm Comparison
Compare different ML algorithms on the same dataset; evaluate selection trade-offs.
AIMX Software — AI for ML Experiments
A powerful, no-code interactive platform covering the entire ML workflow. Train and deploy AI models directly to hardware — no programming required. Two tiers available.
5 apps · 6 experiments · 10 classification · 10 regression · 5 clustering models · Input: ML Kit CSV only
5 apps · Same 6 experiments · 42 classification · 40 regression · 10 clustering models · Accepts external CSV data
| Application | AIMX Standard — Included with MLK | AIMX Advanced — Add-on Purchase |
|---|---|---|
| Experiments App (6 in both) |
1. Advanced Data Audit & Exploration 2. Uncovering Patterns & Relationships in Data 3. Statistical Summarization & Data Aggregation 4. Ranking & Visualizing Feature Importance 5. Data Transformation via Principal Component Analysis 6. Interactive Exploration of a Classification Tree Model |
Same 6 experiments as AIMX Standard |
| Classification App |
10 models: Logistic Regression: Binary GLM, Efficient Logistic Discriminant Analysis: Linear Discriminant Naive Bayes: Gaussian Naive Bayes SVM: Linear SVM · KNN: Fine KNN Decision Trees: Fine Tree Ensemble: Random Forest, Gradient Boosting Neural Network: Trilayered Neural Network |
42 models: Decision Trees: Fine, Medium, Coarse Discriminant: Linear, Quadratic Logistic Regression: Binary GLM, Efficient Logistic Naive Bayes: Gaussian, Kernel SVM: Linear, Quadratic, Cubic, Fine/Medium/Coarse Gaussian (6) Efficient Linear: SGD, Ridge · KNN: 6 variants Kernel Approx.: SVM Kernel, LogReg Kernel Ensemble: Boosted, Bagged, Subspace DA/KNN, Random Forest, Extra Trees, Gradient Boosting, RUSBoosted (8) Neural Network: Narrow, Medium, Wide, Bilayered, Trilayered (5) Modern Boosting: Hist GBM, LightGBM, XGBoost, XGBoost RF (4) |
| Regression App |
10 models: Linear: Linear, Robust Linear, Stepwise Linear Decision Tree: Decision Tree SVM: Linear SVM · KNN: Fine KNN Ensemble: Random Forest Neural Network: Trilayered NN, Bilayered NN, Medium NN |
40 models: Linear/Regularized: Linear, Interactions, Robust, Stepwise, Ridge, Lasso, ElasticNet, Bayesian Ridge (8) Regression Trees: Fine, Medium, Coarse SVM: Linear, Quadratic, Cubic, Fine/Medium/Coarse Gaussian (6) Efficient Linear: SGD, Linear SVM · KNN: 6 variants Ensemble: AdaBoost, GBM, Random Forest, Extra Trees, Hist GBM (5) Gaussian Process: Squared Exp, Matern 5/2, Rational Quadratic (3) Neural Network: Narrow, Medium, Wide, Bilayered, Trilayered (5) Kernel Methods: SVM Kernel (RBF), Least Squares Kernel |
| Clustering App |
5 models: Centroid-Based: KMeans, MiniBatchKMeans Density-Based: DBSCAN Probabilistic: Gaussian Mixture Hierarchical: Agglomerative |
10 models: Centroid-Based: KMeans, MiniBatchKMeans Hierarchical: Birch, Agglomerative Density-Based: DBSCAN, OPTICS, HDBSCAN, MeanShift Graph/Propagation: Affinity Propagation Probabilistic: Gaussian Mixture |
| Deployment App | 2 analyses: Classification & Regression | 2 analyses: Classification & Regression |
| Input Data | ML Kit CSV only (exported via CARIMO front-end software) | Relevant CSV data from external sources |
DC Motor Kit
The gold standard for control systems laboratories — born at IIT Bombay and used in classroom teaching for over 15 years. Affordable, portable, with interactive software and MATLAB/Simulink interface.
Classroom Tested — 15+ Years
Used in control systems teaching at IIT Bombay; battle-tested across India's top engineering institutions.
Make-in-India Initiative
Developed under the MHRD Virtual Labs project; fully aligned with national manufacturing priorities.
Bundle of 7 Interactive Experiments
Transfer function analysis, PI speed control with disturbance rejection, integrator anti-windup, PID position control and more.
Portable & Affordable
Low-cost, highly portable; creates real-world plant experience. Available on Govt. e-Marketplace (GeM).
P, PI, PID control software is included with the kit · MPC software is available as a separate add-on
Downloads & Resources
All documents link directly to Google Drive — always the latest version.
