Robotics Study Guide

📖 Core Concepts Robotics – interdisciplinary field that designs, builds, and operates robots integrating power, mechanics, control, and software. Actuator – “muscle” that turns stored energy (electric, hydraulic, pneumatic, etc.) into motion. Manipulator & End‑effector – robotic arm (manipulator) and its functional tip (gripper, suction cup, etc.). Perception → Processing → Action – three‑stage control loop: sensors gather data, compute decisions, then drive actuators. Kinematics vs. Dynamics – Kinematics maps joint angles ↔ end‑effector pose; Dynamics relates forces/torques ↔ accelerations. Series Elastic Actuation (SEA) – motor + compliant spring to improve force control, safety, and energy efficiency. Swarm Robotics – many simple robots whose global behavior emerges from local interaction rules. --- 📌 Must Remember Forward Kinematics: compute end‑effector pose from joint angles. Inverse Kinematics: compute joint angles for a desired pose (often multiple solutions). Direct Dynamics: \[ \mathbf{a} = \mathbf{M}^{-1}(\mathbf{\tau} - \mathbf{C}\mathbf{\dot q} - \mathbf{g}) \] → accelerations from forces. Inverse Dynamics: find required torques \(\mathbf{\tau}\) for a desired acceleration profile. Zero Moment Point (ZMP): balance condition where the resultant ground reaction force passes through a point inside the support polygon – prevents tipping. Passive‑dynamic walking can be up to 10× more energy‑efficient than active ZMP control. Common Power Sources: wired electricity, batteries (consider safety, cycle life, weight), internal‑combustion generators. Sensor Types & What They Measure: Vision – color/IR images, depth. Lidar – distance via laser time‑of‑flight. Radar – range & velocity via radio waves. Sonar – underwater distance via sound. Tactile arrays – pressure, vibration, temperature. Safety Standards for Cobots: force thresholds and workspace monitoring to protect humans. --- 🔄 Key Processes Robot Perception → Decision → Actuation Cycle Acquire raw sensor data → filter/fuse → compute desired motion → send commands to actuators. Inverse Kinematics Solution (e.g., for a 6‑DOF arm) Define target pose → apply analytical or numerical method → check joint limits & collisions → output joint angles. Series Elastic Actuator Control (Impedance Control) Measure spring deflection → compute force \(F = k \cdot \Delta x\) → adjust motor torque to achieve desired impedance. Swarm Coordination Each robot executes simple rule (e.g., “move toward uncovered area”) → collective coverage emerges. Dynamic Walking (ZMP) Algorithm Predict CoM trajectory → compute foot placement ensuring ZMP stays inside support polygon → adjust gait in real time. --- 🔍 Key Comparisons Electric DC Motor vs. AC Motor – DC: portable, brushed/brushless, common in small robots. AC: high power, dominant in industrial robots & CNC machines. Linear Actuator vs. Rotary Motor + Leadscrew – Linear: direct in/out motion, often hydraulic or pneumatic. Leadscrew: rotary motor converts to linear motion, higher precision. Piezoelectric vs. Ultrasonic Motor – Piezo: nanometer resolution, high‑frequency, limited stroke. Ultrasonic: high torque at low speed, no gearbox needed. Series Elastic vs. Rigid Actuation – SEA adds compliance → better force control & safety; rigid provides higher bandwidth but less safe interaction. ZMP Control vs. Passive‑Dynamic Walking – ZMP: active control, higher energy use. Passive‑dynamic: leverages natural dynamics, far more efficient on flat terrain. --- ⚠️ Common Misunderstandings “All robots need batteries.” – Many industrial robots are tethered to mains power; batteries are chosen for portability, not necessity. “Inverse kinematics always yields a unique solution.” – Redundant manipulators often have multiple feasible joint configurations; choose based on obstacles or joint limits. “Higher torque means faster motion.” – Torque and speed are trade‑offs; high torque at low speed (e.g., ultrasonic motors) may be ideal for precise positioning. “Swarm robots act independently.” – Their global behavior depends on local interaction rules; removing communication can break the emergent function. --- 🧠 Mental Models / Intuition Robot as a Human Body: power source = heart, actuators = muscles, sensors = senses, control system = brain. Series Elastic Actuator: imagine a spring between you and a heavy box; you feel the force through spring compression → you can safely push/pull without hard impact. ZMP: think of a tightrope walker balancing a pole; the pole’s contact point must stay within the foot area to stay upright. Swarm Behavior: like birds flocking – each follows simple “stay close, avoid collisions” rules, producing coordinated flight. --- 🚩 Exceptions & Edge Cases Series Elastic Actuators lose bandwidth at very high frequencies; not ideal for ultra‑fast tasks. Piezoelectric actuators have limited stroke (< 100 µm); unsuitable for large‑range motions. Zero‑Moment‑Point control fails on highly uneven terrain; passive‑dynamic or hybrid approaches are needed. Battery selection: high‑energy‑density cells may have poor cycle life – trade‑off for long‑duration missions (e.g., space rovers). --- 📍 When to Use Which Choose Actuator Type: Need high precision, tiny displacement → Piezoelectric. Need high torque, low speed, compact → Ultrasonic motor. Need compliant, safe interaction → Series Elastic or Air Muscle. Select Locomotion Mode: Indoor flat floor → wheeled (4‑wheel) robot with ZMP control. Rough outdoor terrain → tracked or legged (quadruped) robot with compliant joints. Quiet, soft‑contact tasks → Air muscle or SMA actuators. Control Strategy: Simple, well‑modeled task → Reactive control (sensor → motor mapping). Uncertain environment → Adaptive/fuzzy or ANN‑based control. Multi‑robot coverage → Swarm local‑rule algorithm. --- 👀 Patterns to Recognize “Sensor → Fuse → Model → Command” pattern in most control pipelines. Redundant DOF → multiple inverse‑kinematics solutions → look for joint‑limit or obstacle avoidance clues. High‑frequency vibration in sensor data → likely from piezo or ultrasonic actuators, not environmental noise. ZMP inside support polygon → stable stance; if ZMP drifts to edge → imminent tip‑over. --- 🗂️ Exam Traps Distractor: “Series elastic actuators improve speed.” – They improve force control and safety, not speed. Trap: Assuming “all legged robots use ZMP.” – Many use passive‑dynamic or hybrid methods, especially for efficiency. Misleading choice: Selecting “battery” as the only power source for a large industrial arm – most are wired to mains. Wrong comparison: Equating “piezoelectric actuator” with “large‑stroke linear actuator.” – Piezo offers nanometer precision, not large travel. Confusing sensor ranges: Lidar provides accurate distance up to tens of meters; Radar excels at longer ranges and velocity detection – mixing them up can lead to incorrect design choices.