Haptic designs have become increasingly practical in recent years, particularly for use in remote-control telerobotic systems.
The future is here: It’s now possible to remotely operate equipment while receiving physical and visual feedback from the distant end of a device. This technology — known as telepresence — is useful in harsh or hazardous environments where operators must be kept out of harm’s way, as in nuclear, military, and ultra-high vacuum applications.
Telepresence cues come in the form of instrument readings, including measurement readouts, close-up visual displays of the system’s interaction with the application, and force transmission and feedback. The latter — systems that transmit a simulated feel of remote forces to a human operator — are called haptic systems. These provide essential cues that enable execution of remote tasks using either actual or scaled forces and positions, transferred from physical machinery operations to an operator’s remote station.
Haptics allow for feedback and movement of remote objects over a range of forces that reflects the spectrum of human physical touch and strength. Consider the forces necessary to push something, compared to feeling an object’s texture: These functions invoke disparate forces and sensations.
Pushing an object, for example, exerts forces that can be quite large, observable by the dynamic properties of objects handled in everyday life. Such forces are sensed by stretch receptors in the muscles and ligaments of the body, giving rise to proprioceptive feedback — which in turn gives the body its sense of position and the forces it is applying.
In fact, perception of an object’s characteristics is affected by the dynamics of the object being felt. For example, a human catching a ball can roughly estimate the ball’s weight by the force required to bring it to a stop. Applying a haptic system to similar tasks requires that the system reflect dynamic behavior by accurately following the remote system’s dynamic movements.
On the other hand, the forces associated with the sensing of texture are very low, but observable at a high frequency. When one slides fingers over a tabletop, for example, fingertip nerves can detect small surface ripples at up to 5 kHz.
In general, this range of forces and dynamic responses is too dissimilar to be easily met by one simple actuator system.
A haptic design’s human-machine interface (HMI) is what applies the forces to the body, and its configuration defines both system functionality and operational limitations.
An ideal haptic HMI gives users the sensation that they are directly operating remote equipment, or manually executing operations. In a remotely driven vehicle, for example, a good haptic HMI provides realistic forces to critical controls such as steering.
More challenging is remote object manipulation. Consider the task of picking up a screwdriver, engaging it with a screwhead, and then unscrewing the screw. Nature has provided humans with an impressive range of motion to execute such tasks. To allow remote execution of driving a screw, a haptic HMI would need to apply the forces experienced during this procedure to the operator’s proprioceptive receptors — which would require many actuators positioned near the person’s hands and arms.
Systems developed to apply such forces — such as the ExoHand from Festo (see page 40) and Cyberglove system developed at Stanford University — often resemble specialized exoskeletal devices. However, these are complex and typically cumbersome devices more suited to research than rough-and-tumble operation in industrial environments.
Focused purpose = more practical
More practical designs integrating haptics are typically custom-tailored systems that provide a limited but focused set of forces and movements based on an abstracted version of the tool being used. One example is the da Vinci Surgical System, manufactured by Intuitive Surgical Inc., Sunnyvale, Calif., and used for telerobotic surgery. These da Vinci robotic cells do not send complete force representation to the surgeon, but incorporate a simple grip through which an end-effector manipulator is moved.
In fact, now on the haptic horizon are commercially available systems adapted to machinery with an exchangeable set of tools, each used to perform different operations. Here, the operator would use the haptic HMI to induce tool changes at the remote robot, for a useful increase in reconfigurability.
Desktop and commercial-off-the-shelf type manipulators are low-force devices capable of applying around one to two Newtons of force to the touch. They utilize small direct-drive motors with low inertia. However, they are unsuitable for operations where higher force is required for reliable, intuitive control.
If an operator must steer a vehicle through rough terrain at a reasonable speed, an awareness of the vehicle loads (including traction and lateral accelerations) improves control and mechanical robustness. Forces here are typically high with small fluctuations in force providing realistic, intuitive feedback to the operator.
Geared electric actuators are suitable in this application. However, actuator inertia and friction can preclude the simple feedback arrangements of desktop devices, instead requiring a force sensor or equivalent to sense the system’s forces and eliminate friction to improve dynamic behavior. Systems of this kind can be tailored to meet the special requirements of military and hazardous operations.
For more information, call (888) 534-1222 or visit wittenstein-us.com
Realtime control — wireless considerations
The inclusion of haptics in telepresence systems requires realtime interactive operation, to allow the operator to effectively guide the movement of remotely operated equipment through its environment. Therefore, the data link between the operator and remote system, which is often wireless, must provide for realtime control loops and support robust stability levels. Communication latency and dropouts can be problematic; long-range control (especially those via satellite) can be challenging to implement without significant degradation of dynamic performance. For this reason, current efforts in R&D are aimed at improving wireless functionality for haptics.