Skip to main content
Robotics & Embodied AI

⏱ About 20 min20 XP

Module Check: Making It Move

This lesson recaps and tests the full arc of Module H3: Control, Motion, and Manipulation. You have traveled from the fundamental control problem through kinematics, dynamics, PID, motion planning, locomotion, manipulation, and safe physical interaction. The flashcards below cover the key terms. The quizzes probe understanding at the level of reasoning and application, not just recall. The final activity asks you to synthesize concepts across lessons to solve a novel problem.

Flashcards — click each card to reveal the answer

Module Review Quizzes

A robot arm is commanded to hold a payload at a fixed position. Its position controller reads the joint angles and computes corrective torques. A sensor fails and begins reporting incorrect joint angles. Compared to an open-loop controller running the same task, how does the closed-loop controller's behavior change?

A 6-DOF arm with joint angles (q1...q6) needs to move its end-effector along a straight line in Cartesian space. Why is straight-line Cartesian motion not the same as straight-line motion in joint space, and why does this matter for path planning?

A PID joint controller has Kp = 50, Ki = 0.5, Kd = 8. The arm is moving fast toward a setpoint and overshoots by 4 degrees before the controller reverses direction. Which gain is primarily responsible for reducing this overshoot, and what would happen if it were set to zero?

An RRT planner is run for a 6-DOF arm in a complex cluttered environment. After 10,000 samples it has not found a path to the goal. What are the two most likely explanations, and how would you distinguish between them?

A quadruped robot is walking at low speed using a gait where three feet are always on the ground. The team wants to increase speed significantly. What control transition is required, and what new stability challenge does it create?

A cobot in admittance control is assembling a product when an unexpected human hand touch applies a 40 N force to the robot's end-effector. The admittance law has M = 2 kg, B = 50 N*s/m, K = 100 N/m. At the instant of contact (velocity = 0, position at reference), what is the initial acceleration of the virtual mass, and which direction does the robot move?

Module Synthesis: The Full Stack

  1. This final activity asks you to trace a complete robotics task through every layer of the module's conceptual stack. No design document — just clear, precise answers that demonstrate you can reason across all the lessons.
  2. The task: A 6-DOF cobot arm must pick up a glass from a table (known position, but glass position varies by up to 2 cm) and place it in a dishwasher rack (known position). The cobot operates in a kitchen where humans may be present.
  3. Answer each question in 3-5 sentences:
  4. 1. The Control Problem (Lesson 1): Describe the full sense-decide-actuate loop for this task. What is the plant? What are the setpoints at each phase (reach, grasp, lift, place)? What is the most important disturbance to reject?
  5. 2. Kinematics (Lesson 2): The vision system detects the glass center at position (x=0.42, y=0.15, z=0.31) in the robot's base frame. The gripper must approach from above with fingers horizontal. How does the controller convert this task-space target into joint commands, and why might there be multiple valid joint configurations?
  6. 3. Dynamics (Lesson 3): The glass weighs 0.3 kg. When the arm is extended horizontally at shoulder height, why must the shoulder motor produce more torque than when the arm hangs vertically? Which term in the manipulator equation captures this?
  7. 4. PID Control (Lesson 4): During the lift, the arm uses joint-level PID controllers. The glass is slightly heavier than estimated, causing a 0.8-degree sag in the elbow joint. Which PID term will most effectively correct this without oscillation, and why does the proportional term alone leave a residual error?
  8. 5. Motion Planning (Lesson 5): The dishwasher rack is partially obstructed by a cabinet door. How should the motion planner account for this obstacle, and why is RRT more appropriate here than a pre-taught joint-space trajectory?
  9. 6. Manipulation (Lesson 7): The glass is smooth-sided. Describe the grasp strategy: What end-effector is appropriate? What grip force is needed to support the glass weight given a coefficient of friction of 0.4? (F_grip = weight / (2 * mu)). What sensing would detect slip?
  10. 7. Safe Interaction (Lesson 8): A child reaches toward the glass just as the arm begins moving. What must the cobot's controller do in the next 10 milliseconds? Name the specific sensing method and control response that makes this safe.