Virtual Fixtures

Virtual Fixtures are fundamentally categorized by their behavioral effect on the operator's input.

• Guidance Fixtures act as attractors, channeling the user's commanded motion along a preferred path or toward a target. Examples include a virtual tunnel for guiding a surgical tool or a magnetic line for aligning a component.
• Forbidden-Region Fixtures (or repulsive fixtures) act as constraints, creating virtual boundaries that the controlled system cannot penetrate. These are used to prevent collisions with sensitive anatomy or keep a tool within a safe workspace.

The same underlying geometry, like a virtual plane, can be implemented as either a guidance surface to slide along or a forbidden wall to avoid.

A defining feature is the spatial registration of the virtual geometry with the real environment. The fixture must be accurately aligned with the physical workspace, which requires precise tracking of the robot, tools, and the environment.

• This overlay is typically presented to the operator via a visual display (e.g., an augmented reality headset or monitor) and/or through haptic feedback on the control interface.
• The fixture exists in the task space (the 3D world where the robot operates), not just the control interface, making it an embodied augmentation of the operator's perception and action.

The method of enforcing the fixture's constraints is a critical engineering choice tied to the robot's control architecture.

• Compliant (Impedance) Control: The robot acts as a programmable spring-damper system. When the user commands motion into a forbidden region, the controller generates a virtual force opposing the command, making the interface feel 'stiff' or 'springy'. The user feels the fixture.
• Admittance Control: The controller modifies the user's position command before it is sent to the high-gain position controller. If a command would violate a constraint, it is filtered or redirected. The interface may not provide direct force feedback, but the robot's motion is constrained.

The choice affects transparency, stability, and the required hardware (e.g., force-sensing capability).

Virtual Fixtures are a foundational technology for two primary HRI paradigms:

• Bilateral Teleoperation: Here, fixtures are crucial for stability and performance. They can compensate for time delays in communication links by keeping the slave robot within safe regions, and they provide haptic cues to the master-side operator, enhancing situational awareness.
• Shared Autonomy: In this paradigm, the fixture is a direct manifestation of the robot's autonomous assistance. It dynamically blends human input with machine-generated guidance or constraints to accomplish a task, effectively allocating authority spatially. For example, a fixture may allow free motion in safe areas but enforce strict guidance near a delicate target.

Advanced Virtual Fixtures are not static geometries. They can be dynamic, changing shape, position, or stiffness in real-time based on:

• Sensory Feedback: A forbidden region around a moving organ expands and contracts with its physiological motion.
• Task Phase: A guidance tunnel appears only when the tool needs to be inserted, then disappears.
• User Performance: The stiffness of a guidance fixture may adaptively increase if the user shows tremor or decrease if they demonstrate high skill.

This context-awareness, often powered by real-time perception systems, makes fixtures powerful tools for adaptive assistance.

The utility of Virtual Fixtures manifests in three core objectives:

• Enhanced Precision: By reducing the effective degrees of freedom or filtering tremor, fixtures enable users to perform tasks beyond innate human motor capability, such as microsurgery or nanoscale assembly.
• Guaranteed Safety: Forbidden-region fixtures provide a software-enforced safety layer that prevents catastrophic errors, protecting patients, expensive equipment, or the robot itself from collisions.
• Accelerated Training: Guidance fixtures serve as in-task tutors, allowing novices to learn optimal paths and techniques by physically feeling the correct motion, reducing training time for complex manual tasks.

In nuclear decommissioning, underwater repair, or space operations, Virtual Fixtures enable precise remote manipulation despite communication latency and limited sensory feedback.

• Application: Guiding a robotic arm to turn a specific valve in a cluttered panel by overlaying a snap-to-path fixture that constrains motion to the correct rotational axis.
• Key Mechanism: Admittance Control is often used, where the fixture applies virtual forces to the master controller, making it feel physically harder to deviate from the prescribed path.

In shared-autonomy cobot workflows, Virtual Fixtures assist human workers with high-precision, repetitive tasks like part insertion, welding, or quality inspection.

• Types Used: Attractive Fixtures (e.g., a magnetic-like guide for inserting a peg) and Repulsive Fixtures (e.g., a virtual wall preventing the tool from scratching a finished surface).
• Outcome: Dramatically reduces training time, decreases assembly errors, and increases throughput by blending human dexterity with robotic consistency.

Virtual Fixtures are used in robotic exoskeletons and therapy devices to guide patient movements during motor re-learning after a stroke or spinal cord injury.

• Implementation: A tunnel-in-space fixture can constrain a patient's arm to move along a physiologically correct trajectory during a reaching exercise.
• Adaptive Feature: The fixture's stiffness or guidance level can be dynamically adjusted based on patient performance, providing more assistance when needed and fading it as recovery progresses.

For piloting drones or remote vehicles, Virtual Fixtures simplify complex maneuvers and prevent loss-of-control incidents.

• Use Case: A corridor fixture for a UAV flying through an urban canyon, keeping it centered and preventing collisions with buildings.
• Use Case: An avoidance volume fixture around a fragile object during a drone-based inspection, ensuring the vehicle maintains a safe standoff distance automatically.

Virtual Fixtures are implemented through specific control paradigms that define how they influence the human-robot loop.

• Admittance Control: The robot measures force/torque from the human and moves accordingly; fixtures are implemented as virtual springs and dampers in this mapped workspace. Ideal for large, powerful robots.
• Impedance Control: The robot defines a dynamic relationship between its position and the force it exerts; fixtures modify this relationship to create virtual walls or guides. Common for lightweight, back-drivable arms.
• Overlay Types: Hard Fixtures are absolute constraints (cannot be penetrated). Soft Fixtures are permeable but provide increasing resistance.

A control paradigm where task execution authority is dynamically allocated between a human operator and an autonomous system. Unlike static automation or full manual control, Shared Autonomy blends human intent with machine assistance in real-time.

• Blended Control: The system continuously interprets user inputs (e.g., joystick motions) and combines them with its own autonomous plans to produce the final robot command.
• Intent Inference: Often relies on probabilistic models to infer the human's high-level goal from low-level inputs, allowing the autonomy to assist without explicit commands.
• Application: Critical for complex tasks where human oversight is needed but machine precision or computational speed is beneficial, such as surgical robotics or assisted driving.

A master-slave control scheme for remote manipulation where the operator receives kinesthetic force feedback from the robot's environment. This creates a sense of telepresence, allowing the user to 'feel' remote objects.

• Force Reflection: The slave robot measures interaction forces (e.g., from contact or tool use) and reproduces them on the master controller held by the operator.
• Key Components: Requires a master device (input + force output), a slave manipulator (robot arm with force sensors), and a control law ensuring stability despite communication delays.
• Virtual Fixtures Role: Fixtures are often implemented within this architecture as software constraints that channel the operator's commanded forces or motions, enhancing precision and preventing collisions.

A system design principle enabling dynamic, on-the-fly modification of a robot's level of self-governance. It allows smooth transitions between fully autonomous, semi-autonomous (shared control), and fully manual teleoperation modes.

• Dynamic Mode Switching: The level of autonomy can be changed by the user, triggered by context (e.g., task complexity), or requested by the robot itself (e.g., during uncertainty).
• Interface Requirement: Requires clear mode awareness interfaces so the human always understands who is in control.
• Relation to Fixtures: Virtual Fixtures can be enabled, disabled, or their stiffness adjusted as part of transitioning between autonomy levels. For example, a high-guidance fixture may be used in assisted mode but disabled in full manual mode.

The use of tactile or force feedback to guide a user's physical movements. In robotics, this is often delivered through a force-feedback joystick or exoskeleton.

• Guidance vs. Constraint: While a Virtual Fixture defines a region or path in space, Haptic Guidance is the actuation method used to convey that fixture to the user. It renders attractive or repulsive forces.
• Types of Guidance:
  • Forbidden Region Virtual Fixture (FRVF): Applies a repulsive force if the user commands motion into a restricted zone.
  • Guidance Virtual Fixture (GVF): Applies an attractive force toward a desired path or surface, like a magnetic groove.
• Perceptual Effect: Creates an intuitive, 'hands-on' feeling of being physically guided or blocked, reducing cognitive load.

A reactive navigation and control method where the robot (or its control point) moves under the influence of an artificial potential field. This is a core mathematical formulation for many Virtual Fixtures.

• Field Composition: The total field is a sum of attractive potentials (pulling toward a goal) and repulsive potentials (pushing away from obstacles or restricted regions).
• Implementation for Fixtures: A Virtual Fixture is defined by its potential function. For example, a forbidden region has a high repulsive potential that generates a force opposing user input.
• Advantages & Limitations: Computationally efficient for real-time control but can lead to local minima (points where attractive and repulsive forces cancel, stalling motion). Used extensively in motion planning and hptic rendering.

A robot control paradigm where an applied force (e.g., from a human or the environment) is interpreted as a command for motion. It is the inverse of Impedance Control (where motion is interpreted as a force command).

• Core Law: Force Input → Velocity/Position Output. The controller acts like a virtual spring-damper system: Motion = Admittance * Force.
• Critical for Physical HRI: Makes a robot compliant and safe for direct physical interaction, as it 'gives way' when pushed.
• Connection to Fixtures: Virtual Fixtures are often implemented within an admittance control framework. The 'force' is the sum of the human's input force and the virtual force generated by the fixture's potential field. This combined force is then admitted to create motion, seamlessly blending human intent with software guidance.