A rigid motion is a transformation that moves a shape from one position to another without changing its size or shape. Every distance between points stays exactly the same, and every angle stays exactly the same. The figure you end up with is congruent to the one you started with. In geometry, the term “isometry” means the same thing: a transformation that preserves distance.
The Core Rule: Distance Stays the Same
The single requirement that makes a motion “rigid” is distance preservation. If two points are 5 centimeters apart before the transformation, they must be 5 centimeters apart after it. Side lengths don’t stretch or shrink, interior angles don’t open or close, and the overall proportions of the figure remain identical. Two figures are congruent if and only if a rigid motion can map one onto the other.
This is what separates rigid motions from other geometric transformations. A dilation (scaling a shape up or down) changes distances between points, so it is not rigid. A shear, like pushing the top of a square sideways to make a rhombus, warps angles and lengths, so it also fails the test. If any pair of points ends up closer together or farther apart than it started, the transformation is non-rigid.
The Four Types of Rigid Motion
In a flat plane, every rigid motion is one of four types, or a combination of them.
- Translation (slide): Every point moves the same distance in the same direction. Think of sliding a book across a table without rotating it.
- Rotation (turn): One point, called the center of rotation, stays fixed while every other point swings around it by the same angle. A clock hand rotates around the center of the clock face.
- Reflection (flip): Every point is mirrored perpendicularly across a line, called the line of reflection. The result looks like the original seen in a mirror placed along that line.
- Glide reflection: A translation followed by a reflection across a line parallel to the direction of the slide. Footprints alternating left and right along a straight path are a classic example.
Any rigid motion you can perform on a flat surface will turn out to be one of these four, or can be broken down into a sequence of them. Translations, rotations, and glide reflections keep the “handedness” of a figure (a clockwise-labeled triangle stays clockwise). Reflections reverse it, producing a mirror image.
How Rigid Motion Differs From Other Transformations
The easiest way to spot a non-rigid transformation is to check whether the shape has been stretched, compressed, or distorted. Scaling a triangle so its sides double in length is a dilation. The angles stay the same, but the distances between points all change, so it’s not rigid. Shearing a rectangle into a parallelogram changes both side relationships and angles, so that’s not rigid either.
A rigid motion can reposition a shape anywhere and in any orientation, but it can never make it bigger, smaller, or warped. If you can lay the original figure directly on top of the transformed one and every edge and angle matches perfectly, a rigid motion got you there.
Rigid Motion in Robotics
Rigid motion isn’t just a classroom concept. It shows up whenever engineers need to describe how a solid object moves through space. In robotics, each segment of a robot arm is treated as a rigid body connected by joints that either rotate or slide. Figuring out where the tool at the end of the arm (the “end effector”) winds up after every joint moves is a core problem called kinematics, and it relies entirely on chaining together rigid motions: rotations at each joint and translations along each link.
Positioning a robotic gripper at a precise point in space, predicting whether a robot’s motion will collide with obstacles, and designing mechanisms that move tools from one location to another all depend on rigid motion math. The robot’s “workspace,” the full range of positions and orientations its end effector can reach, is calculated by sweeping rigid motions through every joint’s range.
Rigid Motion in Computer Vision
Computer vision systems use rigid motion to track objects in video. When a camera watches a coffee mug being picked up, the software models the mug as a rigid body and estimates six parameters at every frame: three for translation (how far left/right, up/down, forward/back) and three for rotation (tilting around each axis). Together, those six numbers fully describe any rigid motion in three-dimensional space.
This is sometimes called “pose tracking,” meaning the system continuously identifies both the position and orientation of the object. The rigid body assumption makes the math tractable: instead of tracking every point on the mug independently, the algorithm only needs those six numbers to know exactly where every point on the mug has moved.
Rigid Motion in Molecular Biology
Biochemists use rigid motion when modeling how proteins interact. In a technique called rigid body docking, scientists treat each protein as a solid shape and systematically test how one protein can be rotated and translated to fit against another, much like finding how two puzzle pieces snap together. The software evaluates the fit for all possible translations at once using fast mathematical shortcuts, then tests rotations one by one.
This rigid body approximation works well for a substantial fraction of protein interactions, though it has limits. Real proteins flex and shift shape slightly when they bind, so the models allow a small amount of overlap to account for minor structural changes. When the shape change between a protein’s free-floating form and its bound form is too large, the rigid assumption breaks down and more flexible modeling is needed.
Representing Rigid Motion With Matrices
In two dimensions, any rigid motion combining a rotation by some angle with a translation can be written as a single 3-by-3 matrix that acts on a point’s coordinates. The top-left portion of the matrix handles the rotation, while an extra column holds the translation distances. This compact format lets computers apply a rigid motion to thousands of points with one matrix multiplication per point.
In three dimensions, the same idea extends to a 4-by-4 matrix. A 3-by-3 rotation block sits in the upper left, a translation column sits on the right, and a bottom row of zeros plus a one keeps everything consistent. This “homogeneous coordinate” representation is the standard in computer graphics, robotics, and any field that needs to chain multiple rigid motions together efficiently. Multiplying two of these matrices gives you the single matrix for doing both motions in sequence.