Overview
free

Course Setup and the Incremental Ladder
Why "Motors to Swarms"
How to Use This Course
The Incremental Ladder (Step 0 to Step 7)
The Course Lenses
Diagram Legend and Notation Types

Robots as Perceive–Plan–Act Systems
Sensors to Computation to Actuators
Automation vs Robotics
Robot Taxonomy

Bodies, Joints, and Degrees of Freedom
Links, Joints, and Degrees of Freedom
Joint Types and Mobility Primitives
Common Kinematic Structures and the Boundaries They Impose

Coordinate Frames and Transformations
Reference Frames and Operational Naming
Rigid Transforms Conceptually: Rotation and Translation
Frame Trees: Consistency, Calibration Dependencies, and Error Propagation

Sense–Think–Act Architectures
Layering the Robot: Time-Scale Boundaries Between Control, Perception, Planning, and Task Logic
Real-Time vs Best-Effort Compute: Deadlines, Jitter, and Where Determinism Matters
Single Robot vs Fleet Layers: Adding Supervisory and Coordination Planes

Diagramming Robotic Systems
Kinematic and Frame Diagrams: The Robot's Spatial Contract
Control Loop Diagrams: Inner Loops, Outer Loops, and Time-Scale Separation
Perception, Navigation, and Coordination Graphs: Dependencies as Failure Domains

Step 0 Forward Kinematics
Joint Space vs Task Space: Commanding Joints vs Commanding Poses
Computing Pose from Joints: Forward Kinematics as a Deterministic Geometry Pipeline
Worked Structures: A 2-Link Planar Arm and Simple Mobile Bases

Step 0 Inverse Kinematics (Conceptual)
Pose to Joint Configuration: Why "The Inverse" Is Often Ambiguous or Impossible
Multiple Solutions and Constraints: Preferences, Limits, and Collision Constraints Shaping IK Outputs
Numerical vs Analytical Approaches: Trade-Offs in Generality, Speed, and Robustness

Step 0 Constraints and Workspaces
Joint Limits and Collisions: Constraints as Safety and Feasibility Boundaries
Singularities Conceptually: Where Small Joint Changes Produce Unstable or Unbounded Task-Space Behavior
Workspace and Reachability Diagrams: Making Feasibility Visible Before You Write Control Code

Step 1 Actuation Fundamentals
Actuator Families: DC, BLDC, Servos, Linear Actuators and What They Imply for Control Effort (Conceptual)
Gearboxes and Backdrivability: Torque-Speed Trade-Offs and the Operational Consequences for Safety
Power and Thermal Limits: Why "More Torque" Is Always Coupled to Current, Heat, and Duty Cycle

Step 1 Sensing the Robot and the World
Core Sensors: Encoders, IMUs, Limit Switches, Simple Range Sensors and What Each Can and Cannot Tell You
Sensor Characteristics: Range, Resolution, Noise, Bandwidth as Constraints on Achievable Control
Mounting and Calibration Concepts: Alignment, Biases, and Why Frames Must Match Kinematics

Step 1 Feedback Control: Concepts
Controlled Variable, Setpoint, Error: The Minimal Vocabulary of Closed-Loop Control
Proportional Control and Oscillation: Why Feedback Can Destabilize if You Ignore Dynamics and Delays
PID Intuition: What P/I/D Contribute and How Each Term Fails When Misapplied (Conceptual)

Step 1 Joint and Wheel Controllers
Position, Velocity, Torque Modes: Choosing What the Actuator Tries to Regulate
Cascaded Loops: Inner Velocity and Outer Position and Why Time-Scale Separation Matters
Conceptual Tuning: Stability Versus Responsiveness and the Cost of Aggressive Gains

Step 2 Dynamics (Conceptual)
Mass, Inertia, Friction, Disturbances: Why Robots Are Never Purely Kinematic in Practice
Dynamics as Control Reality: How Unmodeled Dynamics Appear as Tracking Error and Instability
Manipulators vs Mobile Bases: How Morphology Changes the Dominant Dynamics and Constraints

Step 2 Motion Profiles and Trajectories
Velocity and Acceleration Limits - Respecting Actuators and Mechanics as First-Class Constraints
Trapezoidal and S-Curve Profiles - Shaping Motion to Reduce Jerk, Slip, and Oscillation
Trajectories as State Sequences - Feeding Low-Level Loops with Time-Indexed Targets

Step 2 Cascaded Control Architectures
Current/Torque, Velocity, Position Loops: Layering Control to Match Time Scales
Separation of Time Scales: Why Inner-Loop Stability Is Prerequisite for Outer-Loop Intelligence
Robustness Considerations: Disturbance Rejection, Saturation, and How Limits Reshape Stability

Step 2 Safety, Limits, and Interlocks
Soft Limits and Hard Stops: Physical Versus Software Enforcement and Failure Containment
Fault Detection Signals: Overcurrent, Overheating, Stall Detection and What Each Implies About Root Cause
Emergency Stop and Safe States: Designing Behavior for the Worst Second of the Robot's Life

Step 2 Calibration and Homing
Homing at Startup: Establishing a Known State Before Autonomy Begins
Zeroing Encoders and State Initialization: Preventing Drift from Becoming "Truth" in the Stack
Sensor Failure and Fallback: Degraded Modes That Preserve Safety When Calibration Is Suspect

Step 3 Sensor Modalities for Perception
Cameras, Depth, Lidar-Like, Radar: Sensing Choice Determines What Is Observable
2D vs 3D Trade-Offs: Cost, Compute, Robustness, and Failure Modes
Sensor Fusion (High-Level): Reducing Ambiguity While Introducing Coupling

Step 3 Localization Basics
Pose Estimation: What It Means to Know "Where You Are" in a Changing World
Odometry and Drift: Why Dead Reckoning Always Decays and How to Plan for It
External References: Fiducials, Beacons, GNSS-Like Signals, and the Dependency Risks They Introduce

Step 3 Mapping and Environment Representation
Occupancy Grids and Costmaps: Representing Free Space and Risk as Planning Inputs
3D Maps Conceptually: Point Clouds and Voxel Grids and Their Compute and Storage Costs
Semantic Layers: Zones, Rooms, Objects and Why Meaning Helps Task Planning but Complicates Correctness

Step 3 SLAM (Simultaneous Localization and Mapping)
Joint Estimation Conceptually: Why Mapping and Localization Are Coupled Problems
Loop Closure and Drift Correction: How Systems Repair Accumulated Error and What Can Go Wrong
Trade-Offs: Accuracy, Compute Load, Robustness, and the Operational Consequences of Each

Step 3 Perception Pipelines
Raw to Representation: Filtering, Features, and State Estimation as an Evidence Pipeline
Latency vs Fidelity: When "Better Perception" Makes the Robot Worse by Missing Deadlines
Feeding Planning and Control: Defining Interfaces So Perception Is Usable, Testable, and Replaceable

Step 4 Path Planning in Known Maps
Search in Grids and Roadmaps: Planning as Cost-Minimizing Route Finding
Cost Functions: Length, Time, Safety Margins, and How Costs Encode Product Priorities
Global vs Local Planners: Long-Horizon Optimality Versus Short-Horizon Safety

Step 4 Obstacle Avoidance and Local Planning
Reactive vs Planned Avoidance: Speed of Response Versus Predictability
Dynamic Obstacles: Short-Horizon Planning Under Uncertainty and Sensing Gaps
Aggressiveness vs Safety: Tuning Local Planners as Risk Management

Step 4 Navigation Stack Architectures
Pipeline Layers: Perception -> Localization -> Mapping -> Planning -> Control as a Composable Stack
Global vs Local Costmaps: Different Representations for Different Time Horizons
Recovery Behaviors: Stuck Detection, Replanning, and Bounded Failure Handling

Step 4 Task and Behavior-Level Planning (Conceptual)
State Machines, Task Graphs, and Behavior Trees
Sequencing Navigation and Manipulation
Failure Handling and Replanning at the Behavior Level

Step 4 Specialization: Manipulation vs Mobile Navigation
Configuration-Space Planning for Arms
Planning in Clutter
Different Constraints for Bases vs Arms

Step 5 Hardware/Compute Stack for Robots
Low-Level Controllers vs High-Level Compute
MCU, SBC, External Compute
Power Distribution and Compute Reliability

Step 5 Real-Time Constraints in Robotics
Hard vs Soft Real-Time
Scheduling Loops, Periods, Deadlines, and Jitter
Balancing Perception, Planning, and Control

Step 5 Communication Middleware and Messaging
Message Buses and Pub/Sub: Designing Internal Robot Communication as a Distributed System
Topics, Services, Actions: Request/Response and Event Patterns for Robotics Workflows
Bandwidth and Latency on the Robot Network: When Internal Networking Becomes the Bottleneck

Step 5 Control Stack Architectures
Stack of Stacks: Firmware, Mid-Level Control, High-Level Autonomy and Where Interfaces Belong
Safety Loop Separation: Ensuring Safety-Critical Behaviors Are Not Coupled to Best-Effort Autonomy
Safe Software Upgrades: Updating Deployed Robots Without Destabilizing Control or Safety Invariants

Step 5 Fault Detection, Watchdogs, and Recovery
Watchdogs and Heartbeats: Detecting Stalled Compute and Unresponsive Subsystems
Fault Trees and Graceful Degradation: Designing Degraded Modes That Preserve Safety and Partial Utility
Failsafe States and Safe-Stop: Choosing Stop Behaviors That Remain Safe Under Sensor or Compute Uncertainty

Step 6 HRI Fundamentals
Human Roles: Operator, Supervisor, Co-Worker, Bystander and How Each Changes Interface Obligations
Trust, Predictability, Transparency: Designing Behavior People Can Safely Interpret
Interaction Zones: Personal, Social, Public and Why Distance Is Part of the Interface

Step 6 Modes of Interaction
Teleoperation, Shared Control, and Autonomy: Selecting Where Control Authority Lives
Visual Interfaces: Dashboards, UIs, and AR Overlays as Supervision Tools
Physical Interaction: Hand-Guiding and Force-Limited Behaviors as Safety and UX Design

Step 6 Safety Around Humans
Speed and Separation Monitoring: Bounding Motion Risk with Sensing and Policy
Collaborative Behaviors and Safeguards: What "Collaboration" Requires Beyond Proximity
Signaling Intent: Lights, Sounds, and Motion Cues as Safety Interfaces

Step 6 Command and Configuration Interfaces
Goal and Constraint Interfaces: Specifying Tasks Without Forcing Users Into Low-Level Control
Conflicting Commands and Priorities: Arbitration, Authority, and Safe Refusal Behavior
Operator Workflows: Simple, Robust Sequences for Routine Operations and Incident Handling

Step 6 Explainability and Debuggability for Operators
What Is It Doing and Why: Exposing Autonomy State in Ways Operators Can Act On
Logs and Visualizations: Introspection Surfaces That Support Field Diagnosis
UX for Misbehavior: Designing Operator Tools for Containment, Recovery, and Learning

Step 7 Multi-Robot Coordination Basics
Centralized vs Decentralized Coordination: Control Authority and Failure-Domain Trade-Offs
Shared Maps and Shared Queues: Consistency, Staleness, and What "Shared Truth" Means for a Fleet
Communication Topology and Reliability: Coordination Under Delay, Loss, and Partial Connectivity

Step 7 Task Allocation and Fleet Management
Allocation Mechanisms: Auctions, Queues, Policies as Different Fairness and Optimality Postures
Balancing Load and Priorities: Travel Time, Urgency, and Robot Capability Constraints
Fleet Dashboards and Controls: Operational Interfaces for Supervisors, Maintenance, and Incident Response

Step 7 Swarm-Inspired Behaviors (Conceptual)
Local Rules to Emergence: Why Small Policies Can Produce Large-Scale Behavior
Formations, Flocking, Coverage: Pattern Families and Their Robustness Properties (High-Level)
Scalability and Failure: What Swarms Tolerate Well and What They Amplify

Step 7 Cloud and Edge Integration
Offloading Compute: Mapping, Global Planning, Analytics, and the Coupling Risks to Connectivity
Cloud-Backed Telemetry and Updates: Observability, Logs, and Fleet-Wide Change Management
Online Learning and Model Distribution: Conceptual Capabilities and the Safety Risks They Introduce

Step 7 Productizing Robotic Systems
Prototype to Product: Industrial Design, Manufacturability, and the Operational Cost of Physical Choices
Reliability, Lifecycle, Maintenance: Spares, Wear-Out, and Field Service as System Requirements
Regulatory and Deployment Context: Operating Environments, Approvals, and Constraints (High-Level)

Step 7 Reference Architectures for Robotic Systems
Case Architectures: Warehouse Robot, Delivery Robot, Manipulator Cell, Drone Fleet (Conceptual)
Mapping Steps to Product Maturity: What Changes When You Leave the Lab and Scale Deployments
Case-Style Trade-Offs: Representative Boundary Decisions and Their Consequences

Control and Architecture Patterns
Standard Control Architectures: Cascaded Loops and Hierarchical Control as Reusable Motifs
Internal System Architecture Patterns: Tiered Control and Perception-Planning-Control Pipelines
Structural Choices and Failure Domains: How Internal Decomposition Determines What Can Fail Together

Perception and Mapping Patterns
Navigation Pattern Families: Grid-Based Navigation, Visual-Inertial Localization, Laser-Based Mapping (Conceptual)
Hybrid and Layered Maps: Combining Representations to Balance Compute Cost and Robustness
Simplify vs Model Richly: Choosing the Minimum World Model That Supports Safe Behavior

Safety, Reliability, and Validation Patterns
Safety Envelopes and Virtual Fences: Bounding Behavior With Explicit Constraints
Simulation, Real-World, Hardware-in-the-Loop: Test Strategies Across Increasing Realism
Shadow Mode and Phased Rollouts: Validating Autonomy Changes Without Immediate Full-Risk Exposure

Design Checklists for New Robotic Systems
Mechanical and Kinematic Readiness: Geometry, Limits, and Calibration Prerequisites
Sensing, Perception, Planning, Control Readiness: Interfaces, Latencies, and Safety Constraints
HRI and Fleet Readiness: Human Workflows, Operational Tooling, and Multi-Robot Coordination Boundaries

/motors-to-swarms/motors-to-swarms

Dynamic Obstacles: Short-Horizon Planning Under Uncertainty and Sensing Gaps

Sign in to access this lesson.

Sign in Create account