Advanced

By now you're a pretty decent robotics engineer. But if you want to go into R&D, you'll need to pick up more specialized material. This part ties much more closely to your particular research direction, so I can't spell it all out one piece at a time; I can only sketch the route for you.

I'd also strongly suggest getting a copy of the Springer Handbook of Robotics^[3]. When you step into a new area, find the relevant chapter in the Handbook, use it to get the basic outline, and use the references it provides to fill in the gaps quickly.

Three main threads

Let me first make the scope of this part clear. If the introductory part was about the traditional "teach-and-playback" arm (how to get a robot moving, and moving well), then the advanced part is about giving the robot some capacity for autonomous decision-making, so it can cope with a changing environment: perception-based planning and control (what the industry here calls 规控, "plan-and-control"). It's organized around three main threads, one chapter each:

• The mathematical language — Modern Robotics: the space of orientations and poses is not Euclidean; the language of screws, the product of exponentials, and groups clears up the questions the introductory part left open, all at once, and hands you two handy weapons (orientation interpolation and constraint-space construction);
• Perception — 3D Vision: connecting the "see → locate → act" chain, with the focus on calibration and pose estimation, ending on a closed-loop visual-servoing case study;
• Decision-making — Autonomous Planning: moving from teaching to autonomy — configuration space, the map of algorithms, planning under constraints, and the reality of getting a planner onto the production line.

The closing Advanced Practice chapter pairs these three with a hands-on checklist. The three chapters can be read in order, or dipped into as needed.