Domain: Robotics, Mechatronics, and Control Systems Engineering Persona: Senior Robotics Systems Architect
Step 2: Summarize (Strict Objectivity)
Abstract:
This project details the development of an "AutoClub," a mechatronic mini-golf putter designed to autonomously correct for human swing error and hit moving targets. The system utilizes a high-torque brushless DC (BLDC) motor driving a putter head through a drive shaft and a non-backdriveable worm gearbox. Key engineering challenges addressed include the neutralization of reaction torque through motor inertia tuning, 6D trajectory prediction using an Optitrack infrared motion capture system, and the mitigation of software latency between Python-based control logic and hardware. While the prototype successfully executes straight-line putts and tracks moving holes, the current iteration faces limitations in complex physics modeling (sloped terrain and bank shots), which are slated for future software updates.
Technical Breakdown and Key Takeaways:
0:00 Project Objective: The development of a closed-loop control system for a golf club that adjusts the face angle in real-time during a ~200ms human swing to guarantee the ball reaches the target.
1:41 Mechanical Configuration: To avoid ground interference during rotation, the design pivots the club head around a vertical axis. A drive shaft connects a top-mounted motor to a worm gearbox at the base.
3:24 Transmission & Stability: A worm gearbox is employed to prevent the impact of the ball from back-driving the motor. To eliminate gear backlash, a tensioned elastic element is utilized between the head and internal gearing.
4:12 Reaction Torque Compensation: Initial designs used an external counterweight to cancel twisting forces. After identifying that motor acceleration also generates significant torque, the engineer pivoted to using "inertia rings" on the motor to perfectly cancel the putter head's reaction force.
5:52 Component Selection: The system utilizes an ODrive controller for the BLDC motor, 5-axis machined aluminum for the putter head, and SLS (Selective Laser Sintering) 3D-printed nylon for structural housing.
10:14 Environmental Constraints: Modular testing holes were constructed using MDF and steel tubing. Unexpected warping in the MDF (1cm center drop) introduced non-linear ball paths, complicating trajectory calculations.
12:19 Motion Tracking Architecture: An array of 12 Optitrack cameras tracks infrared markers on the club and hole at 240Hz. A custom reflective-tape coating allows the golf ball to be tracked as a single large marker.
15:18 Control Algorithms: The software fits a 6D trajectory to the club’s path, predicting impact speed and orientation while adjusting the motor position to compensate for shaky human wrists.
17:16 Latency Mitigation: Inconsistencies in the Windows/Python stack caused a 10ms+ delay, leading to control loop oscillations (overshooting). This was addressed by a temporal synchronization "hack" that aligns camera data with historical motor position data.
21:04 Dynamic Targeting: The system successfully demonstrated "ball-seeking" capabilities by hitting a moving hole target via real-time intercept prediction.
22:01 Future Iterations: Current limitations include the high computational cost of full physics simulations (currently 25 hours per shot for bank shots). Future updates aim to optimize these simulations for real-time play.
Step 3: Domain-Specific Review
Reviewing Group:Peer Review Panel for Mechatronics & Feedback Control Systems
Panel Summary:
"The 'AutoClub' prototype represents a sophisticated exercise in high-speed system integration. From a mechanical standpoint, the shift from discrete counterweights to tuned motor inertia for torque cancellation is an elegant solution to Newton’s Third Law constraints in a handheld form factor. However, the reliance on a Python/Windows architecture introduces significant deterministic jitter, which the developer mitigated through historical data buffering rather than a true Real-Time Operating System (RTOS).
The sensing layer—utilizing a 240Hz Optitrack array—provides sufficient spatial resolution for 6D pose estimation, but the system's current bottleneck is the 'Integration Hell' of environmental variables. Specifically, the unmodeled bowl-effect of the putting surface and the impulse-induced 'jump' of the ball upon wall impact remain significant variables that the current PID/prediction logic cannot fully resolve without the proposed Part 2 physics engine. The hardware is over-engineered for straight-line accuracy, but the software must bridge the gap between idealized geometry and the high-entropy physics of a non-flat mini-golf environment."