This build was part of the European ROBOMINERS project. The challenge was to achieve high-resolution subsurface mapping to guide autonomous drilling robots. The AR4 was used to precisely position and hold an electrical resistivity tomography sensor against a mine wall. This allowed the team to measure electrical resistivity changes at close range, providing detailed data on the boundaries and composition of mineral deposits hidden behind the rock surface.

Learn More: https://robominers.eu/2023/10/25/european-union-funded-mining-robot-achieves-remarkable-success-in-slovenia-field-trials/

i was inspired by the project and i wanted to build an robot to implement in the factory where i work, but the new manager was not so convinced and i kept working on robot. at first was 3d printed and after i was able to manufacture the robot from aluminium.

I wanted a little experience

an AR4 robotic arm into a precision soldering system using an MK8 extruder with pneumatic coupler for consistent solder feeding. The wrist-mounted design maintains TCP accuracy during material changes, achieving professional-grade solder joints at 2mm/s linear speed with 1 arc/s rotation.

Transitioning from CNC/3D printing to robotics, the AR4MK2 was redesigned to overcome EU sourcing challenges and improve structural integrity. This build replaces imperial bearings with metric components, reduces part count through 3D-printed consolidations, and reinforces critical joints – all while maintaining full AR4 software compatibility.

This AR4-based “HandyBot” understands voice commands to autonomously pick/place tabletop objects using cutting-edge AI vision (Grounding DINO, Segment Anything) and OpenAI Whisper speech recognition. The $2300 system combines an AR4 arm, RealSense D435 camera, and custom ROS 2 pipeline for real-world object interaction.

🔗 Build it yourself: GitHub Repo
💬 Discuss the ROS driver: Community Thread

The AR4 robotic arm system received significant motion control upgrades. The core innovation involves a split-axis homing sequence that prevents potential self-collision during initialization by separating the homing process into two distinct phases.

Motion control improvements center around advanced acceleration algorithms implementing S-curve profiles, replacing the previous linear motion approach. This upgrade enables smooth transitions between waypoints while maintaining positioning accuracy. The Raspberry Pi-based control system executes these motion profiles while managing real-time position feedback.

Operational Workflow

1. System initialization with split homing sequence

2. Automatic centering procedure completion

3. Waypoint execution with dynamic speed adjustment

4. Continuous collision monitoring during operation

5. Graceful error recovery protocols

Performance Outcomes

• 72% reduction in motion-induced vibration

• 100% prevention of homing collisions

• Sub-millimeter positioning repeatability

The complete motion control solution demonstrates how firmware and control logic enhancements can significantly upgrade robotic system capabilities without mechanical modifications. Video documentation showcases the dramatic improvement in motion quality and operational safety.

Goal: Replace manual part handling with automated systems for efficiency/consistency.

Challenges: Retrofitting old drilling machine with Arduino-controlled pneumatics.

Results:

CNC machine: Reliable 15-minute cycles over 3 months.

Drilling machine: 40-second cycles with zero downtime in first week.

Long-Term Testing: Monitoring performance under 8-hour daily operation.