Tutorials

Files are for what already happened. Streaming is for what is happening now: a robot on the bench, a test running in the next room, a process emitting values you want to watch as curves rather than as scrolling numbers. PlotJuggler has streamed since the PJ3 era — its documented streamers cover WebSocket, MQTT, ZeroMQ…

Every tutorial in this series ends the same way — “save the layout” — and this one explains why that advice matters. A layout is PlotJuggler’s saved form of a view: which plots you built, which signals sit on them, and how the view is arranged, captured to a file you can reload instead of…

Built-in transforms handle the everyday jobs — a derivative here, a moving average there. The moment your question becomes “…and now combine these signals in a way the toolbox does not offer”, you need a custom function. PlotJuggler’s Custom Function Editor provides exactly that: you write a small function in Lua that takes multiple recorded…

Every plot so far has put time on the horizontal axis. An XY plot asks a different question: not “how did this signal evolve?” but “how do these two signals relate?” — position against position, output against input, one joint against another. XY plots are a long-standing PlotJuggler feature in PJ3, and PJ4 keeps them…

Two signals, two clocks, two start buttons pressed a second apart — and a plot that insists your systems disagreed when they actually agreed perfectly, just not simultaneously. Time offset is PlotJuggler’s answer: shift a curve (in PJ3) or a whole dataset (in PJ4) along the time axis until events that happened together line up…

One recording tells you what happened. Two recordings — before and after a fix, run A against run B — tell you whether anything actually improved. Comparing multiple bag files is one of the most searched PlotJuggler workflows, and it is also where the two generations differ most clearly: PlotJuggler 4 explicitly supports loading several…

For ROS users, PlotJuggler’s real test is not CSV — it is the recordings the robots already produce: ROS 1 .bag files, ROS 2 bags, and MCAP files. Both PlotJuggler lines open them, but by different routes, and confusing the routes is the main source of “it will not open my bag” frustration. This tutorial…

CSV is where most PlotJuggler journeys begin: it is the format every logger, spreadsheet and test rig can produce, and both PlotJuggler lines read it — CSV is a core loader in PJ3 and stays on the supported file list in PJ4 alongside MCAP, Parquet and PX4 ULog. This tutorial covers the one concept that…

Ten minutes from now you will have a real signal on a real plot — and, more importantly, you will understand the handful of ideas every other PlotJuggler workflow is built from: loading data, dragging signals onto a plot, navigating the timeline, and saving the view as a layout. This walkthrough works for both current…

PlotJuggler has always been extensible — PJ3’s C++ plugins added data loaders, streamers, message parsers and toolboxes — but finding and installing those plugins was a manual, expert-only exercise. PlotJuggler 4 gives the ecosystem a front door: the Extensions Marketplace, built into the application (File → Marketplace), backed by a public plugin registry and a…