Technologies and Terms

Welcome to Technologies and Terms, your ultimate glossary of innovation on Robot Streets. This is where clarity meets curiosity—where complex ideas become clear, and the language of robotics transforms from jargon to inspiration. From AI frameworks and motion algorithms to control architectures, fabrication techniques, and sensor technologies, this section breaks down the concepts that keep the world of robots in motion. Each article bridges theory and application, explaining not just what a term means but why it matters. Whether it’s the neural logic behind machine learning, the precision of SLAM navigation, or the mechanics of robotic perception, you’ll find the insight to connect every component of modern automation. Whether you’re an engineer sharpening your vocabulary or a newcomer exploring the foundations of robotics, Technologies and Terms helps you speak the language of tomorrow’s machines—fluently, confidently, and with imagination. Step inside, and start decoding the world where technology and terminology collide.

1. V–I–P: Volts push, Amps flow, Watts = Volts × Amps—size traces and fuses from peak draw.

2. Logic levels: 3.3V vs 5V CMOS/TTL; use level shifters for safe cross-talk.

3. Pull-up/down: resistors define default states; essential for buttons and open-drain buses.

4. PWM: duty cycle controls average power for motors, LEDs, and heaters.

5. ADC/DAC: bridge analog worlds—filter noise with RC and keep reference clean.

6. UART, I²C, SPI: serial point-to-point; shared address bus; fast multi-line—pick per need.

7. Debounce: hardware RC or firmware timing removes switch chatter.

8. Bridges & drivers: H-bridge for DC, half/full-bridge for coils; add flyback diodes.

9. Grounding: star grounds reduce loops; separate power/logic returns then join once.

10. Protection: ESD diodes, TVS clamps, and polyfuses guard fragile inputs.

1. SLAM = Simultaneous Localization and Mapping: building a map while finding yourself in it.

2. Kinematics vs dynamics: geometry of motion vs forces that cause it.

3. FK/IK: forward maps joints→pose; inverse maps pose→joints (watch for singularities).

4. Pose = position + orientation; often expressed as SE(3) transform.

5. Odometry drift: tiny errors accumulate—loop closure fixes the map.

6. PID: P reduces error, I kills offset, D damps overshoot; add feedforward for snap.

7. Sensor fusion: Kalman/particle filters combine noisy sources into clean estimates.

8. ROS graph: topics (pub/sub), services (RPC), actions (goals with feedback).

9. QoS: reliability and history settings tailor message delivery to real-time needs.

10. PL/SIL: performance/ safety integrity levels quantify risk reduction.

1. Middleware: ROS 2 for modular nodes, logging, and simulation hooks.

2. Simulation: Gazebo/Isaac-style sims for physics + sensor pipelines.

3. Visualization: RViz/Foxglove for TF trees, point clouds, and bag replay.

4. Data capture: rosbag2 + SQLite/Parquet for structured analysis.

5. Controls lab: notebook + plotting libs for PID/MPC tuning sessions.

6. CI/Lint: colcon test, formatters, static analysis for safe merges.

7. Docs: Sphinx/MkDocs with glossaries for shared terminology.

8. Diagramming: PlantUML/SysML for architectures, sequence flows, and statecharts.

9. Instrumentation: USB logic analyzer, oscilloscope, and CAN/Serial analyzers.

10. Versioning: git with conventional commits and CHANGELOG for term updates.

1. State space: ẋ = Ax + Bu, y = Cx + Du—compact modeling for controllers.

2. Observability/controllability: can you infer states and drive them anywhere?

3. Planning: graph search (A*), sampling (RRT*), optimization (MPC) per constraints.

4. Splines & polynomials: jerk-limited S-curves make payload motion smooth.

5. Time sync: PTP/chrony align clocks for multi-sensor fusion.

6. Real-time: deadlines, jitter, and priority inversion—design for worst case.

7. Frames: TF trees track coordinate transforms across moving parts.

8. Safety: watchdogs, heartbeats, and fault states for graceful degrade.

9. Redundancy: voting sensors/CPUs handle outliers and single-point failures.

10. TCO thinking: performance + maintenance + training beats list price alone.

1. “Robot” traces to Czech robota (“compulsory work”) from a 1920s play.

2. “Android” originally meant humanlike form; “gynoid” specifies feminine design.

3. “Cobot” highlights safe human–robot collaboration with limited forces.

4. “Gimbal lock” isn’t a robot bug—it's Euler-angle geometry biting back.

5. “Teach pendant” survives even as modern arms learn from demonstration.

6. “Perception stack” spans lens → ISP → detector → tracker → pose.

7. “Digital twin” = live-synced model that mirrors the physical robot.

8. “Backdrivable” actuators feel light because friction and gearing are tuned down.

9. “Dead reckoning” works—until wheel slip or drift says otherwise.

10. “Edge AI” moves inference on-board to cut latency and bandwidth.

Q: ROS vs ROS 2?
A: ROS 2 adds real-time-friendly DDS, better security, and multi-robot QoS.

Q: LiDAR or camera for SLAM?
A: LiDAR excels in geometry; cameras add rich features—fusion wins.

Q: Visual odometry vs V-SLAM?
A: VO estimates motion only; V-SLAM also builds the map.

Q: Occupancy grid vs mesh?
A: Grids are 2D probabilistic maps; meshes capture 3D surfaces.

Q: Open vs closed loop?
A: Open drives a command; closed uses feedback to hit targets.

Q: TTL vs RS-232?
A: TTL is logic-level; RS-232 is higher-voltage, longer-distance serial.

Q: Can 2D LiDAR map 3D?
A: With motion tilts or multiple layers; otherwise you get slices.

Q: What is TF?
A: A system to track coordinate transforms between robot parts over time.

Q: Kinematics vs dynamics in control?
A: Kinematics sets geometric paths; dynamics ensures forces achieve them.

Q: Homogeneous transforms?
A: 4×4 matrices that combine rotation and translation in one op.

View Product Reviews

RobotStreets

News Street Network

Powered by RedHawks Media

Social