Collaborative Robot (Cobot)

The foundational hardware characteristic of a cobot is its inherently safe design. This includes:

• Force-limiting joints and lightweight structures to minimize kinetic energy.
• Rounded edges and padded surfaces to reduce injury risk during incidental contact.
• Backdrivable actuators that allow a human to easily move the arm when powered on.

This physical design ensures that, even in a fault condition, the robot's basic mechanics present a lower risk profile than a traditional industrial arm.

Power and Force Limiting (PFL) is the core collaborative safety mode defined by the ISO/TS 15066 standard. In PFL mode, the robot's control system actively monitors and restricts the power, force, and pressure exerted at the end-effector and along the arm.

• It uses integrated torque sensors at each joint to measure applied forces.
• The system ensures forces remain below defined biomechanical limits for quasi-static (pinching) and transient (impact) contact.
• This allows for contact to occur without causing pain or injury, enabling true physical collaboration like co-assembly.

A Safety-Rated Monitored Stop is a collaborative operation where the robot ceases all motion as soon as a human is detected entering a predefined collaborative workspace.

• Detection is typically via safety-rated light curtains, laser scanners, or area sensors.
• The robot remains powered and holds position, resuming its task automatically and immediately once the human exits the zone.
• This mode maximizes productivity for tasks where human and robot work sequentially, not simultaneously, within the same cell.

Hand Guiding is an intuitive programming and operation mode where a human operator physically grasps the robot's end-effector or arm and manually moves it through a desired task path.

• The robot's control system enters a zero-gravity or compliant mode, using its actuators to compensate for gravity, making the arm feel weightless.
• The taught positions, orientations, and paths are recorded directly into the robot's program.
• This eliminates the need for complex joystick or teach pendant programming, enabling rapid task setup and redeployment by shop-floor workers.

In Speed and Separation Monitoring (SSM) mode, the robot and human can work in close proximity, but the control system ensures a protective separation distance is always maintained.

• External safety-rated vision systems or lidar track the human's position and velocity in real-time.
• The robot's speed is dynamically adjusted—slowing down as the human approaches and stopping completely if the minimum separation distance is breached.
• This allows for high-speed autonomous operation when the human is far away, with gracefully degrading performance as collaboration becomes closer.

Beyond safety, a key feature enabling collaboration is ease of integration and programming. Cobots are designed as flexible automation tools.

• They often feature drag-and-drop programming interfaces and support high-level languages.
• Universal tool and part adapters (like ISO/ANSI flange mounts) allow quick end-effector changes.
• Built-in force-torque sensing enables advanced applications like precision insertion, polishing, and quality inspection without additional hardware.
• This flexibility allows them to be redeployed for new tasks in hours, not days, adapting to mixed-model and low-volume production.

Cobots automate the loading and unloading of parts from CNC machines, injection molding presses, and stamping presses. This application is ideal for high-mix, low-volume production where flexibility is key.

• Key Drivers: Addresses labor shortages for dull, repetitive tasks and enables lights-out manufacturing.
• Technical Features: Often utilizes vision systems for part localization and force sensing for delicate part handling.
• Example: A UR10e cobot equipped with a pneumatic gripper services three different CNC mills, managing a batch of 50 unique parts with quick changeover between jobs.

Cobots are used to pick products from a conveyor or workstation and place them into cartons, cases, or onto pallets according to a defined pattern.

• Key Drivers: Reduces physical strain on workers from repetitive lifting and bending, while increasing throughput and consistency.
• Technical Features: Requires path planning for collision-free motion in confined spaces and often integrates conveyor tracking for dynamic pick-and-place.
• Example: In a bakery, a FANUC CRX-10iA/L cobot with a vacuum gripper gently places fragile pastries into individual slots in a cardboard box at a rate of 30 units per minute.

Cobots perform precise, repetitive inspection tasks such as visual checks, measurement with probes, or functional testing (e.g., button pressing).

• Key Drivers: Eliminates human error and fatigue in detailed inspection, ensuring consistent quality standards.
• Technical Features: Integrates 2D/3D vision cameras, laser scanners, or force-torque sensors to gather inspection data. The cobot's repeatability (<0.1mm) is critical.
• Example: An ABB YuMi dual-arm cobot uses one arm to hold a smartphone and the other to manipulate a test probe, executing a suite of 50 functional tests on every device off the assembly line.

Cobots handle small parts assembly, insert components, and drive screws with consistent torque. This is common in electronics, automotive sub-assemblies, and consumer goods.

• Key Drivers: Increases production speed, ensures precise torque application, and frees human workers for more complex assembly stages.
• Technical Features: Relies on high-precision tool changers, compliant tooling for part mating, and integrated screw feeders. Force control is essential for successful insertions.
• Example: A Techman TM AI Cobot assembles a power tool, picking gears and housings from kits, applying threadlocker, and driving eight M4 screws to a specified 3.5 N·m torque.

Cobots execute processes that require consistent path and parameter control, such as applying adhesives, sealants, solder paste, or performing light welding (e.g., MIG, laser).

• Key Drivers: Delivers superior consistency in bead quality or material deposition compared to manual operation, reducing rework and material waste.
• Technical Features: Uses process-specific end-effectors (dispense valves, welding torches) and path programming often taught via kinesthetic teaching. May include seam tracking for welding.
• Example: A KUKA LBR iiwa cobot performs a continuous silicone bead seal around an automotive headlight housing, maintaining a perfect 2mm bead width and 1mm height for 500 units per shift.

In non-industrial settings, cobots handle sensitive tasks like pipetting, sample sorting, vial handling, and assisting in surgical or rehabilitation contexts.

• Key Drivers: Provides sterile, tireless assistance for repetitive lab work; enables precise, tremor-free motion in clinical settings.
• Technical Features: Often feature cleanroom-compatible designs, smooth surfaces for easy decontamination, and exceptionally gentle force limiting (<5N).
• Example: In a pharmaceutical lab, a Yaskawa HC10 cobot, sterilized with UV light, transfers trays of PCR plates between incubators, centrifuges, and liquid handling stations overnight.

Hand Guiding is an intuitive cobot operation mode where a human operator physically grasps the robot's end-effector or arm and manually moves it through a desired task path. The robot's control system enters a zero-gravity or compliant state, using its joint torque sensors to detect the applied human force and providing minimal resistance. This mode is essential for kinesthetic teaching and rapid, codeless task programming.

• Primary Use Case: Programming by Demonstration (PbD) for tasks like welding paths, assembly sequences, or inspection points.
• Technical Implementation: Relies on joint torque sensing or current-based force estimation to interpret user intent.
• Safety Integration: Often combined with PFL modes and requires an enabling device (e.g., a deadman switch on the tool) to activate.

A Safety-Rated Monitored Stop is a collaborative safety function where the robot ceases all motion the instant a human is detected entering a predefined collaborative workspace. Unlike an emergency stop, the robot remains powered and ready, holding its position. Motion automatically resumes once the human exits the zone. This function enables sequential collaboration, where human and robot work on a shared workpiece in turns, maximizing space and asset utilization.

• Key Differentiator: Maintains power and program state; no restart procedure required.
• Sensor Dependency: Typically triggered by safety-rated light curtains, laser scanners, or area scanners (e.g., SICK, Banner).
• Application: Ideal for machine tending applications where a human loads/unloads a part, then the robot performs a machining or processing operation.

Shared Autonomy is a control paradigm that dynamically blends human input with robotic automation. Instead of binary manual or fully autonomous modes, authority is continuously shared. The robot interprets high-level human intent (via joystick, gesture, or force) and autonomously handles low-level stability, precision, or constraint satisfaction. This is critical for complex tasks like surgical robotics or assisted assembly.

• Core Principle: Intent Recognition + Autonomous Assistance. The human provides guidance; the robot handles details.
• Implementation Methods: Often uses virtual fixtures (software-defined guidance geometries), admittance control, or model predictive control (MPC).
• Benefit: Reduces operator cognitive load and fatigue while leveraging human judgment and robot precision.

Speed and Separation Monitoring (SSM) is a collaborative safety mode where the robot's speed is dynamically controlled based on the measured distance to a human operator. Safety-rated sensors (e.g., 2D/3D LiDAR) continuously monitor the separation distance. As a human approaches, the robot slows down; if a minimum protective separation distance is breached, the robot stops. This allows for fluid, close-proximity work without physical contact.

• Dynamic Control: Robot velocity is a function of distance, human speed, and system reaction time.
• Protective Separation Distance (PSD): Calculated using formulas in ISO/TS 15066, factoring in robot stopping time, sensor detection latency, and human intrusion speed.
• Application: Used in logistics (AMRs working near people) and large workcell applications where full PFL is not practical.