An inertial reference frame is one in which Newton's laws hold without modification—free bodies maintain constant velocity and net forces cause acceleration proportional to mass. All frames moving at constant velocity relative to an inertial frame are themselves inertial; accelerating frames require fictitious forces for Newton's laws to apply.
From Newton's second law, you know that F = ma describes the relationship between force and acceleration. But acceleration *relative to what*? The answer is: relative to an inertial reference frame — a frame in which Newton's laws hold exactly as written. This is not a circular definition; it is an empirical claim about certain special frames that exist in nature. The distant stars, for practical purposes, define one such frame. Earth's surface is approximately inertial for most engineering problems, because Earth's rotation and orbital acceleration are small enough to be negligible.
The key property of inertial frames is their equivalence under constant-velocity transformation. If you are on a train moving at steady 100 km/h and you drop a ball, it falls straight down from your perspective — exactly as if the train were stationary. From the platform, the ball traces a parabola, but both observers agree on the forces and on the acceleration *relative to their own frame*. This is Galilean relativity: the laws of mechanics are the same in all frames connected by constant-velocity transformations. No mechanical experiment can distinguish between being at rest and moving at constant velocity — only relative motion matters, not absolute motion. Your prerequisite study of relative motion between frames gives you the mathematical machinery to translate between these equivalent descriptions.
The distinction becomes sharp when the frame accelerates. Imagine you are in a car that brakes suddenly: you feel pushed forward, but there is no physical agent doing the pushing. From the ground (an inertial frame), your body simply continues at its original velocity while the car decelerates around you — Newton's first law. From the car's frame (non-inertial), you experience a fictitious force directed forward. This fictitious force is real enough to throw you against your seatbelt, but it has no source; it is entirely an artifact of the accelerating frame. To use Newton's second law in a non-inertial frame, you must add these fictitious terms explicitly: for a frame with acceleration a_frame, you add −m·a_frame as a fictitious force on every particle.
Rotating frames introduce additional fictitious forces beyond simple linear acceleration — most importantly the Coriolis force and centrifugal force. These are essential in geophysical problems (explaining why hurricanes rotate) and appear in rotating machinery analysis. For most undergraduate dynamics problems, however, the key discipline is simpler: identify whether your chosen frame is inertial before writing Newton's second law. Using the lab frame (fixed to Earth) or any frame translating at constant velocity relative to it: you're safe. Using a frame that accelerates or rotates: you must add the appropriate fictitious forces, or transform to an inertial frame first.
This concept builds directly toward Lagrangian mechanics. One of Lagrangian mechanics' elegant advantages is that the equations of motion derived from energy principles are valid in *any* coordinate system — including rotating and accelerating frames — without needing to explicitly identify and add fictitious forces. Understanding why Newton's laws require an inertial frame is precisely what motivates the more powerful Lagrangian formulation.