cs.ROAug 24, 2026

Reproducible Vision-Guided 6-DoF Robotic Manipulator with a Mixed Stepper-Driver Architecture and Browser-Native Control

Authors: Lasan Perera, Deneth Priyadarshana, Dulana Pitiwaduge, Isitha Dinujaya, Mokshan Colambage

Organizations: Department of Electronic and Telecommunication Engineering, University of Moratuwa, Moratuwa, Sri Lanka

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

We present the NeuralNexus Arm, an open, low-cost 6-DOF robotic manipulator built by an undergraduate engineering team, together with the design decisions and debugging experience needed to reproduce it. The arm is driven by a single STM32H743 microcontroller on a custom printed circuit board (PCB) and combines two stepper-driver strategies on one controller: push-pull 3.3 V step/direction outputs for onboard TMC2209 drivers on the three wrist joints, and open-drain outputs for external CL57T and DM542 drivers on the three high-torque proximal joints. We describe the mechanical design, mixed-driver electronics, interrupt-driven firmware, a MATLAB/Simscape-based inverse-kinematics pipeline, a browser-native control interface using the Web Serial API, and a lightweight vision pipeline for object localisation and autonomous pick-and-place tasks. We also document non-obvious hardware and firmware failure modes encountered during the transition from a development board to the custom PCB as reproducibility guidance. All design files and firmware are released openly. The platform actuates all six axes under coordinated control at a 2 kHz update rate and executes both manual and pre-recorded motions from the browser interface.

Figures & tables

RobotCost ($)DOFControllerMotorKey Difference from NeuralNexus
HELENE<1160<11606ESP32 + ROS HostStepperSingle closed-loop stepper driver family throughout (no mixed architecture); ROS-based control, not browser-native; the only competitor with a published ISO 9283 repeatability validation.
iArm≈250\approx 2506Raspberry Pi 4 + ROSSerial Bus ServoServo-actuated, not stepper-driven; far lower torque/payload class; ROS-based host software rather than a zero-install browser interface.
EduSCARA≈150\approx 1504STM32 Nucleo + Python APIHobby ServoOnly 4-DOF SCARA kinematics (RRPR), not a full 6-DOF articulated arm; potentiometer feedback rather than closed-loop stepper/encoder joints.
AR4 (MK3/MK4)≈2000\approx 20006Teensy 4.1 + Arduino Nano/MegaOpen-Loop StepperUses one driver family only (DM542T/DM320T, same open-loop class at different current ratings) – not a mixed architecture; encoders exist but closed-loop correction is not implemented in firmware; Arduino/ROS control, no browser-native UI; non-commercial license restricts redistribution/resale.
PAROL6≳350\gtrsim 3506Custom STM32 Board + Python Commander GUIStepper (Closed-Loop Upgradable)Single driver family on its custom board (closed-loop is an optional FOC-driver upgrade, not a factory heterogeneous design); control is via a desktop Python app, not a browser; closest analog to our custom-PCB approach but without driver-family mixing.
Forte<215<2156Off-the-Shelf Drivers, No Custom BoardStepper (Capstan/Belt Drive)No custom controller PCB at all – uses off-the-shelf drivers with a host PC; achieves low cost via a capstan-cable/belt drivetrain rather than gearbox+belt reductions; open-source release status unconfirmed.
TABLE I : Comparison of NeuralNexus with representative open-source robotic manipulators.
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Fig. 2 : The assembled NeuralNexus Arm.
Fig. 3 : Custom STM32H743 controller PCB, showing the onboard TMC2209 drivers for the wrist joints and the terminals to the external closed-loop drivers.
JointMotor classDriverGPIO mode
J1 (base)NEMA 23DM542Open-drain
J2 (shoulder)NEMA 24CL57TOpen-drain
J3 (elbow)NEMA 24CL57TOpen-drain
J4 (wrist 1)NEMA 17TMC2209 (onboard)Push–pull
J5 (wrist 2)NEMA 17TMC2209 (onboard)Push–pull
J6 (wrist 3)NEMA 17TMC2209 (onboard)Push–pull
TABLE II : Per-joint actuator and driver mapping.
Fig. 4 : Controller architecture. A single STM32H743 drives two stepper-driver families over separate electrical interfaces: push–pull 3.3 V logic to the onboard TMC2209 wrist drivers, and open-drain, common-anode 5 V signalling to the opto-isolated external closed-loop drivers on the proximal joints. Per-joint AS5047P magnetic encoders return over SPI [ 15 , 19 , 17 , 16 , 20 ] .
Motor classJoint(s)Supply voltage
NEMA 17J4–J6 (wrist)12 V
NEMA 23J1 (base)24 V
NEMA 24J2, J3 (shoulder, elbow)48 V
5 V (opto)J1–J3 driver inputs5 V (buck from 12 V)
Logic (MCU)—3.3 V (LDO)
TABLE III : Motor-supply rails by motor class.
JointTranslation (m)Joint AxisType
J1(0,  0.017,  0)(0,\;0.017,\;0)(0,0,−1)(0,0,-1)Revolute
J2(0,  −0.075,  −0.138)(0,\;-0.075,\;-0.138)(0,0,1)(0,0,1)Revolute
J3(0.310,  0,  −0.024)(0.310,\;0,\;-0.024)(0,0,1)(0,0,1)Revolute
J4(−0.060,  0.160,  −0.051)(-0.060,\;0.160,\;-0.051)(0,0,1)(0,0,1)Revolute
J5(0.041,  0,  −0.1065)(0.041,\;0,\;-0.1065)(0,0,1)(0,0,1)Revolute
J6(0,  0.110,  −0.041)(0,\;0.110,\;-0.041)(0,0,1)(0,0,1)Revolute
TABLE IV : Joint frame transformations extracted from the MATLAB rigidBodyTree model.
Fig. 5 : Kinematic modelling and motion execution workflow. The manipulator is modelled using a MATLAB rigidBodyTree , inverse kinematics generates the required joint angles, and the computed commands are transmitted to the STM32H743 controller for synchronized stepper motor actuation.
Fig. 6 : Browser-native control panel (Web Serial API).
Fig. 7 : Detected 8×58\times 5 inner-corner patterns from nine calibration images captured at different positions and orientations.
Fig. 8 : Detected object-processing stages used for target localisation.
Fig. 9 : Visualisation of planar homography calibration using the 32 mm32~\mathrm{mm} chessboard. The detected image-plane corner points are mapped to a top-down representation of the calibrated workspace.
SubsystemCost ($)
Actuation — motors, drivers, gripper$300.38
Control, PCB & sensing electronics$252.36
Vision compute$178.80
Mechanical parts (gears, pulleys, belts, bearings)$296.82
Local hardware — misc. components, fasteners$149.96
Enclosure / structural fabrication$293.08
TABLE V : Bill of materials summary by subsystem. Full itemised breakdown in Appendix Acknowledgment .
Fig. 10 : Selected frames from Video 2, showing a full pick-and-place sequence: https://youtube.com/shorts/Vg3pZLFrEdM .
MetricValue
Positioning repeatability±1.3\pm 1.3 mm
Pick-and-place cycle time12.512.5 s
Task success rate97%97\%
Maximum payload20002000 g
Maximum joint speed10∘10^{\circ} /s
TABLE VI : Experimental Performance Summary
ItemQtyUnit ($)Total ($)SupplierUsed in
Actuation — motors, drivers, gripper
NEMA 23 stepper motor (Local)1$22.65$22.65Local ShopJ1 (base)
NEMA 17 stepper motor (Local)3$7.45$22.35AliExpressWrist joints
NEMA 24 stepper motors CL57T drivers+ gearboxes + PSU (StepperOnline)2$119.20$238.40Stepper OnlineJ2, J3
DM542 stepper driver (Leadshine, DM542)1$9.83$9.83Local ShopJ1
Gripper / end-effector hardware (3D Printed)——$7.15Local ShopGripper
TABLE VII : Detailed bill of materials. Costs converted from LKR at 1 LKR = $0.00298 (Aug. 2026).
More in this topic

Robotics

Robotics studies the design and control of machines that perceive, move, and act in physical environments.

See all 755 papers

Explore similar work

CardsList

  1. Nori Bot: A Sub-$1,000 Floor-to-Counter Mobile Manipulator

    May 15, 2026Antonio Li, Sungjoon Park, Wen Ni ChewRobotic ControlMobile Manipulation

    Department of Computer Science, Columbia University

  2. Cutting the Cord: System Architecture for Low-Cost, GPU-Accelerated Bimanual Mobile Manipulation

    Mar 10, 2026Artemis Shaw, Chen Liu, Justin Costa +5RoboticsMobile Manipulation

  3. EgoThumb: A Vision-Driven Supernumerary Robotic Thumb for Autonomous Grasp and Holding Assistance

    Oct 27, 2026Jianyu Hou, Jiawei Fu, Weikang Wan +3Robotic Hand DesignAssistive Robotics

    Department of Computer Science and Engineering, University of California San Diego, La Jolla, CA 92093, USA. · Sudo AI GmbH, 8050 Zurich, Switzerland.