INDRBT / Reading room
Project purpose and alignment with the original brief
Adapted for browser reading from the project’s illustrated documentation. Numeric results and engineering scope are unchanged. Reproduction notes are optional; the robot and board viewers run from the navigation above.
On this page
INDRBT explores one connected engineering problem: choose an original robot geometry, use the same geometry to plan a useful tool motion, estimate its loads and gripper requirements, and demonstrate feedback control in a reproducible digital model. The intended result is a credible student engineering project that a professor or reviewer can inspect, run and question.
The design is a six-revolute-axis arm with an interchangeable parallel gripper. Its nominal shoulder-centred reach is 790 mm, its assumed payload target is 4 kg, its repeatability target is ±0.1 mm and its jaw opening is 90 mm. These choices sit within the ranges in the original brief. The design is original; commercial robot compatibility and physically measured performance are not claimed.
Follow one demonstration through the engineering
- Inspect the CAD component catalog, dimensioned interfaces and six poses. The mounting, joint, tool and gripper interfaces explain how the geometry supports the proposed mechanism.
- Read the frame and kinematic formulation. The central SI parameters drive the transform chain and CAD dimensions. The browser viewer uses meshes from those actual solids.
- Run the pick-and-place task. MOVE J handles joint transfers, regular approximate MOVE L segments handle approach and retreat, and the gripper sequences a rigid ideal attachment. Position and orientation follow the mathematical tool pose.
- Compare loads, structure and actuator sizing with the PID model and the five parameter studies. The studies explain effects and tradeoffs using generated data.
- Inspect the electronics architecture, component/BOM guide, native schematic, routed PCB and actual check reports. This is the low-voltage joint/sensor interface portion of the hypothetical controller architecture.
- Read the executed validationSource file and requirements matrix. Their scope, source bindings and failure conditions matter as much as the illustrations.
Original goals and their concrete project form
| Original brief | Project form | Where to assess it |
|---|---|---|
| Original, realistic six-axis mechanical design and interchangeable gripper | Reproducible Build123d solids, native FreeCAD assembly, STEP/STL parts, joint frames, guides, mounting and service interfaces | CAD methodology, catalog, drawings and native model |
| Engineering reasons for dimensions and choices | Central specification, original flange, lumped mass assumptions, opposed-screw gripper and explicit tradeoffs | Requirements, decisions, mechanics |
| Correct mathematics and model-driven motion | FK, bounded numerical IK, Jacobian, seeded workspace, constrained planning and recorded task | Theory guide, kinematics, trajectory data and browser checks |
| Simplified digital feedback control | Saturated PID with antiwindup and independent inertia/damping plants; step and tracking errors | Architecture, control evidence |
| Meaningful conceptual electronics | Regulated logic supply, encoder/limit inputs, external-drive command interface, RS-485, native PCB and manufacturing exports | Electrical chapter, native PCB evidence |
| Educational/research value | Link-ratio, payload, duration, controller-gain and finger-length sweeps | Experiments, KPIs, engineering report |
| Reproducible, tested and organized delivery | Installable package, regression tests, native checks, source-bound reports, numbered folders and offline viewing | Setup, completion checklist, validation and organization manifests |
| A credible engineering portfolio with an authentic process | Clear README, drawings, actual CAD renders, demonstrations, assumptions, decisions and recorded corrections | Engineering report, journal, Git history and releases |
Completion means the documented digital brief
For the ordinary demonstration, follow the source README.
python scripts/run_validation.py --full additionally rebuilds the primary
CAD and numerical outputs, then refreshes native FreeCAD, Blender, drawings/render
composition and sampled clearance before checking the full project. It requires
installed FreeCADCmd, Blender and KiCad alongside the Python dependencies.
Tool-path overrides are documented in CAD reproduction
and electronics reproduction.
Fresh browser and isolated-wheel evidence are required whenever their bound
artifact or package sources change; the validator rejects stale evidence.
The original request explicitly excludes physical construction. It permits analytical beams and simplified joint plants, and asks for representative interference screening and a meaningful conceptual controller subsection. Accordingly, the implemented scope includes these models with stated limits. Hardware payload/repeatability, fatigue, EMC, functional safety, complete manufacturing tolerances, production firmware, frictional grasp physics and continuous swept collision certification remain unverified.
The current status must come from executed checks that match the current sources. The final review records defects found and corrected, while the scope audit maps the detailed brief to evidence. A declaration of completion does not replace those checks or turn assumptions into measured performance.