Robotic Arms and Manipulators: Mechanics, Kinematics, and Applications

Anatomy of a Robotic Arm

Every robotic arm consists of the same fundamental components: a base, a series of rigid links connected by joints, and an end effector at the working tip. The base is the fixed foundation, typically bolted to a floor, ceiling, wall, or mobile platform. Links are the rigid structural members, usually made of cast aluminum or steel. Joints connect consecutive links and provide the arm's mobility.

The two main joint types are revolute joints, which rotate around an axis (like your elbow or shoulder), and prismatic joints, which slide along a linear axis (like a drawer rail). The vast majority of industrial robotic arms use only revolute joints because rotary motion is mechanically simpler, more compact, and easier to seal against dust and fluids than linear motion.

Degrees of Freedom

The number of independent movements a robotic arm can make is called its degrees of freedom (DOF). To position an object at any point in 3D space with any orientation requires six DOF: three for translation (X, Y, Z position) and three for rotation (roll, pitch, yaw). This is why the standard industrial robot arm has six revolute joints. Each joint provides one degree of freedom, and together they allow the end effector to reach any position and orientation within the arm's workspace.

Seven-DOF arms add a redundant joint that allows the arm to reach the same end-effector position in multiple elbow configurations. This redundancy is valuable for reaching around obstacles, avoiding joint limits, and optimizing criteria like minimizing energy consumption or maximizing manipulability. The KUKA LBR iiwa and Kinova Gen3 are popular seven-DOF arms used in research and collaborative applications.

Arms with fewer than six DOF sacrifice either positional or orientational flexibility. A four-DOF SCARA arm can position its tool anywhere in a horizontal plane and raise or lower it, but it cannot tilt the tool away from vertical. This is perfectly adequate for pick-and-place and insertion tasks where the object is always approached from above.

Kinematics: The Mathematics of Arm Motion

Kinematics describes the relationship between joint angles and the position of the end effector without considering the forces involved.

Forward kinematics answers: "If I set the joints to these angles, where will the end effector be?" This is computed using the Denavit-Hartenberg (DH) convention, a systematic method for assigning coordinate frames to each joint and multiplying 4x4 transformation matrices to get the final position and orientation. Forward kinematics always has a unique solution and is computationally inexpensive.

Inverse kinematics (IK) answers the harder question: "If I want the end effector at this position and orientation, what joint angles do I need?" IK is the fundamental computation in robot arm programming because tasks are naturally specified in the tool's coordinate frame ("put the welder here, pointed this way") but must be executed in joint space ("set joint 1 to 45 degrees, joint 2 to 90 degrees, and so on").

IK is challenging for several reasons. For a six-DOF arm, there can be up to 16 different joint configurations that all place the end effector at the same target. The planner must choose among these solutions based on criteria like proximity to the current configuration, avoidance of joint limits, and avoidance of singularities (configurations where the arm loses effective degrees of freedom). Some targets may have no IK solution at all because they are outside the arm's reachable workspace.

Analytical IK solutions exist for many standard arm geometries and provide exact, fast answers. For non-standard geometries or redundant arms, numerical IK methods (such as the Jacobian pseudoinverse or optimization-based solvers) iteratively converge on a solution. Libraries like KDL (Kinematics and Dynamics Library), IKFast, and TRAC-IK implement these solvers and are integrated into the MoveIt motion planning framework.

Dynamics: Forces and Torques

While kinematics describes geometry and motion, dynamics describes the forces required to produce that motion. The dynamic model computes how much torque each motor must generate to move the arm along a desired trajectory, accounting for:

Gravity: The arm must continuously exert torque to hold itself up against gravity. The shoulder joint of a long, heavy arm might need to produce hundreds of newton-meters of torque just to maintain its position.

Inertia: Accelerating the links requires torque proportional to their mass and the square of their distance from the joint axis. This is why robot arms taper (get thinner) toward the end effector, as mass near the tip has a disproportionate effect on the torque required at the base joints.

Coriolis and centrifugal effects: When multiple joints move simultaneously, the motion of each joint creates forces on all other joints. These coupling effects become significant at high speeds and must be compensated for accurate trajectory tracking.

Friction: Gear reducers, bearings, and seals all produce friction that opposes motion. Friction models range from simple Coulomb friction (a constant opposing force) to more sophisticated models that account for velocity-dependent viscous friction and static friction (stiction).

Accurate dynamic models enable feedforward control, where the controller precomputes the expected motor torques based on the planned trajectory and the dynamic model, then adds feedback corrections on top. This approach achieves much better trajectory tracking than feedback alone, because the feedforward component handles the predictable physics while feedback handles only the unpredictable disturbances.

End Effectors: Tools of the Trade

The end effector is the business end of a robotic arm, the device that actually interacts with the workpiece. The right end effector determines whether a task succeeds or fails.

Parallel jaw grippers use two flat fingers that move toward each other to clamp an object. They are simple, reliable, and adequate for objects with flat parallel surfaces. Payload capacity ranges from grams (for electronics assembly) to hundreds of kilograms (for automotive parts).

Vacuum grippers use suction cups connected to a vacuum pump to pick up objects with flat, smooth surfaces. They are the standard choice for handling sheet materials (glass, metal panels, cardboard), boxes, and bags. Multiple suction cups can be arranged in arrays to handle large or irregularly shaped objects.

Magnetic grippers use permanent magnets or electromagnets to pick up ferromagnetic (steel, iron) objects. Electromagnets have the advantage of release on command, simply turning off the current drops the part. They are common in steel fabrication and automotive stamping operations.

Dexterous hands have multiple articulated fingers capable of grasping objects of various shapes and sizes. The Shadow Dexterous Hand has 20 degrees of freedom across five fingers and approaches human-level manipulation capability. However, dexterous hands are mechanically complex, expensive ($100,000 or more), and require sophisticated control algorithms. Research groups at Google, Meta, and universities are using reinforcement learning to teach dexterous hands to manipulate objects through trial and error in simulation.

Tool changers allow a robot arm to automatically swap between different end effectors during operation. A robot might use a vacuum gripper to pick up a flat panel, set it down, switch to a screw driver to fasten it, then switch to a camera for inspection, all without human intervention. ATI Industrial Automation, Schunk, and Staubli manufacture robotic tool changers.

Applications Across Industries

Manufacturing: Welding, painting, assembly, machine tending, packaging, palletizing, and quality inspection account for the majority of robotic arm installations. The automotive industry alone uses an estimated 1 million robotic arms globally.

Surgery: The da Vinci Surgical System from Intuitive Surgical uses miniature robotic arms inserted through small incisions, controlled by a surgeon sitting at a console. The robot filters out hand tremors, scales down motions for precision, and provides 3D magnified visualization. Over 12 million surgical procedures have been performed with da Vinci systems worldwide.

Space: The Canadarm2 on the International Space Station is a 17-meter robotic arm with seven joints that captures visiting spacecraft, moves equipment, and supports astronaut spacewalks. It operates in the vacuum of space, withstanding extreme temperature swings from minus 100 to plus 200 degrees Celsius.

Laboratory automation: Robotic arms in pharmaceutical and biotech labs handle thousands of samples per day, performing pipetting, plate handling, centrifuging, and sample storage. Lab robots enable high-throughput screening of drug candidates, processing volumes that would be impossible manually.

Food service: Robotic arms now prepare meals in fast-food restaurants, brew coffee in airports, and mix cocktails in bars. Miso Robotics' Flippy arm works in restaurant kitchens, frying chicken and flipping burgers with consistent timing and temperature control.