DIY Robot Lamp Full-Body Motion Test: A Firsthand Account from Components to Coordinated Movement

A Dancing Lamp: A Milestone in a DIY Robot Project

Within the robotics hobbyist community, a DIY project blending mechanical engineering, motor control, and creative design has been drawing widespread attention. A maker shared a video on Reddit of the first full-body motion test of their homemade robot lamp—although the movements were still a bit clumsy and the cabling was a chaotic mess, all the joints finally moved in coordination. Naturally, the creator then had it "break into a dance."

This seemingly simple "dancing" demonstration actually marks a critical turning point for a complex electromechanical project, transitioning from component assembly to system integration and tuning. For anyone who has worked on robotics development, getting all the joints to move in coordination for the first time is an achievement worth celebrating in its own right.

Project Hardware Selection: A Deep Dive into Robstride Actuators

According to the creator's update notes, the project uses Robstride series actuators, specifically the 00 series and the Edulite 05 models. There's clear engineering logic behind this hardware choice.

The Positioning and Advantages of Robstride Actuators

Robstride is an integrated joint actuator brand that has gradually gained popularity in the open-source robotics and humanoid robot space in recent years. Such actuators typically integrate a brushless motor, reducer, encoder, and drive controller, allowing them to directly output high-torque rotational motion while supporting multiple control modes such as position, velocity, and torque.

Robstride was developed by a team in Shenzhen, China, and its design philosophy is heavily influenced by the Quasi-Direct Drive technology approach from MIT's Mini Cheetah project. Quasi-direct drive technology was first systematically validated around 2018 by Benjamin Katz and Jared Di Carlo at MIT's Biomimetics Lab through the Mini Cheetah project. Its core idea is to abandon the high-reduction-ratio (50:1 to 100:1) harmonic drive schemes used in traditional robot joints, instead adopting low-reduction-ratio (3:1 to 10:1) planetary gearboxes paired with high-torque-density outrunner brushless motors. The biggest advantage of this choice is back-drivability—externally applied forces can be sensed by the motor through the gearbox, enabling impedance control and torque control, not just position control. Traditional high-reduction systems, due to excessive gear friction, are almost impossible to back-drive, resulting in robots that lack compliance in contact tasks and are prone to damaging their own structures or injuring people during collisions.

Unlike traditional servo motors, Robstride uses low-reduction-ratio planetary gearboxes (typically between 6:1 and 9:1), which allows it to retain a degree of torque amplification while still offering high back-drivability and torque transparency—that is, external forces can be transmitted backward and sensed, which is crucial for robot joints requiring compliant control.

Its CAN bus communication interface also makes multi-joint cascade control relatively straightforward, making it ideal for makers to quickly build prototype systems. The CAN (Controller Area Network) bus was originally developed by Bosch in 1983 for automotive electronic systems. Its differential signaling mechanism, multi-master arbitration protocol, and hardware-level error detection make it naturally suited to robotic systems with complex electrical noise. In multi-joint robotics applications, the CAN bus allows a main controller to connect up to 127 slave nodes in a single-bus daisy chain, typically at a communication rate of 1 Mbps with latency as low as under 1 millisecond. Compared to traditional PWM-controlled servos or UART serial ports, the CAN bus can simultaneously transmit multidimensional state feedback such as position, velocity, and current, and possesses inherent broadcast capabilities, making it well-suited to motion control scenarios requiring real-time synchronization of multiple joints.

In this project, the 00 series handles the lightweight motion of the lamp's arm—it's small and cost-effective—while the Edulite 05 drives the base or main joint, providing greater torque output. This "large and small combination" selection approach reflects the creator's practical consideration of the load distribution across the joints.

Why Integrated Actuators Have Become the Maker's First Choice

The popularity of integrated joint actuators can be traced back to the quasi-direct drive scheme validated around 2015 by the MIT Biomimetics Lab in its Cheetah robot project. Subsequently, products like the Dynamixel servo series, the ODrive open-source driver, the T-Motor AK series, and China's Unitree A1 motor emerged, integrating what were previously separate motors, drivers, encoders, and reducers into plug-and-play modules. Behind this trend lies the substantial decline in the cost of permanent magnet materials, power semiconductors, and position sensors, as well as the proliferation of industrial buses like CAN/EtherCAT in consumer-grade products.

For the maker community, this means that the hardware cost of building a multi-degree-of-freedom robotic arm has dropped from tens of thousands of dollars a decade ago to just a few thousand RMB. In the past, building a multi-DOF robot joint system had an extremely high barrier to entry, requiring one to design motor drive circuits and debug encoder feedback from scratch. Today, the widespread availability of integrated actuators has dramatically lowered the difficulty of developing robotic arms, robot dogs, and even humanoid robots. Makers can now devote more energy to motion planning and overall design, rather than getting bogged down in low-level driver debugging.

From "It Moves" to "It Moves Gracefully": Three Core Technical Challenges

The creator candidly admitted that the current motion is still quite "rough" and outlined three improvement directions for the next phase. These three points happen to represent the core challenges that carry a robotics project from prototype to maturity.

Smooth Trajectory Planning

The first motion test is often just a simple point-to-point movement, with each joint independently reaching its target position, which easily results in noticeable jerkiness. To achieve smoother motion, trajectory interpolation algorithms need to be introduced.

Trajectory planning is one of the core aspects of robot motion control. Its essence is to generate time-sequenced reference commands for the joints while satisfying physical constraints (maximum velocity, acceleration, jerk). There are two levels of planning strategy here: Joint Space Planning directly interpolates the change in each joint's angle over time—it's computationally simple and free of singularity issues—while Cartesian Space Planning first defines the end-effector path in three-dimensional space, then maps each point along the path back to joint angles via inverse kinematics (IK), guaranteeing the geometric accuracy of the tool path.

At the specific algorithm level, the Trapezoidal Velocity Profile is the most basic implementation, dividing motion into three phases: acceleration, constant velocity, and deceleration. The S-Curve further eliminates impacts at velocity inflection points by limiting jerk, while methods such as B-splines or Bézier curves are often used for multi-joint coordinated Cartesian space planning. For performance-oriented applications like a robot lamp, joint-space S-curve interpolation is sufficient to achieve smooth dance movements. These methods make joints more compliant during acceleration and deceleration phases, avoiding mechanical shock while imbuing the overall motion with a stronger sense of life—and for a "dancing lamp," the sense of life in the movement is arguably even more critical than precision.

PID Control Parameter Tuning

"Control tuning" usually refers to the fine adjustment of PID controller parameters. Although the PID (Proportional-Integral-Derivative) controller is a classic algorithm dating back to the 1930s, it remains the most widely used joint position control scheme in industrial and consumer robotics to this day. Its core idea is to calculate the control output through a linear combination of the current error (P term), the accumulated historical error (I term), and the rate of change of the error (D term).

For robot joints, the difficulty in tuning lies in the fact that each joint's moment of inertia, friction characteristics, and load center of gravity change dynamically with the configuration—parameters tuned statically may not remain stable across the entire workspace. When an actuator tracks a target position, improperly set gains can lead to problems such as overshoot, oscillation, or sluggish response.

Of course, there are more advanced control schemes beyond PID. Computed Torque Control is based on the robot's full dynamics model, eliminating nonlinear terms through feedforward compensation, linearizing the system, and then applying PD control—theoretically achieving consistent performance across the entire workspace. Model Predictive Control (MPC) optimizes the control sequence over a future time window within a receding horizon, can explicitly handle constraints such as joint torque and velocity, and has achieved notable success in the real-time motion control of quadruped robots in recent years. But for a lightweight maker project like a robot lamp, adding feedforward torque compensation on top of standard PID is usually the most cost-effective improvement path—by calculating the static support torque of each joint through kinematic analysis and adding it as a feedforward term, motion quality can be significantly improved without introducing a complex dynamics model. Tuning parameters individually for the load characteristics of each joint is the essential path to making robot motion both fast and stable, and it is also the aspect that most tests a developer's patience.

Engineering Cable Management

Seemingly trivial, cable management is actually an important reflection of engineering integrity. The current "tangled mess" of wiring not only affects aesthetics but may also cause cable wear, tangling, and even breakage as joints move repeatedly. A professional wiring scheme must comprehensively consider the joint's range of motion, routing paths, and fixing methods—common techniques include using spiral wrap sleeving to protect dynamic routing zones, reserving sufficient cable slack near the joint's rotation center, and using cable carriers (drag chains) to constrain the direction of harness movement. The completeness of cable management is often an intuitive standard for judging whether a robot project has the potential for mass production or long-term use.

Why This DIY Robot Project Is Worth Watching

A robot lamp easily brings to mind Pixar's classic hopping lamp figure. Pixar's iconic logo character, Luxo Jr. (the hopping lamp), has been a symbolic icon of animated anthropomorphism since the 1986 short film. From an engineering perspective, the reason Luxo Jr. conveys such a strong sense of life to the audience lies in the fact that its motion design strictly follows principles from the twelve principles of animation, such as "Squash and Stretch," "Anticipation," and "Follow Through and Overlapping Action"—the physical essence of these principles is an artistic simulation of momentum conservation, elastic deformation, and inertial effects in real biological motion. Modern robot motion design is now drawing on these animation principles: when designing Boston Dynamics' Atlas robot's dance moves, engineers explicitly added effects such as "anticipation" and "overshoot-and-rebound" into the trajectories. For a DIY robot lamp, after achieving basic trajectory smoothing, the next creative challenge is precisely how to encode these animation principles that make mechanical motion "come alive" into the motion control program.

Giving fluid dynamic expression to a previously static household object is, in itself, a perfect fusion of mechanical art and engineering technology.

From a broader perspective, this project reflects an important trend in current robotics development: the continuous lowering of hardware barriers is unleashing the creativity of individual makers. With integrated actuators and mature control frameworks, independent developers can now build multi-degree-of-freedom motion systems that were once achievable only by professional labs.

The significance of the milestone "all joints are finally moving" should not be underestimated—it means the three major subsystems of mechanical structure, electrical connection, and control logic have been successfully connected for the first time. The subsequent trajectory optimization, parameter tuning, and cable organization are all refinements built upon this foundation.

Conclusion: First Make It Move, Then Make It Graceful

From a pile of parts to a dancing robot lamp, this DIY project showcases the complete arc of robotics development. The creator's pragmatic attitude—not hiding the current roughness and mess while clearly planning the next steps—is precisely the most valuable engineering spirit in hardware development.

For enthusiasts aspiring to build robots, this case offers a practical reference path: choose the right integrated actuators to lower the entry barrier, prioritize getting the whole workflow running, then iterate and optimize step by step. From the democratization of quasi-direct drive technology to open-source implementations of trajectory planning algorithms, the entire tech stack is opening up to individual developers at an unprecedented pace. Here's hoping that, after completing trajectory smoothing and control tuning, this lamp can truly perform a graceful dance—a mechanical ballet blending engineering precision with an animator's soul.