Understanding the factors that influence sensor performance, such as magnet size, elastomer hardness, and elastomer diameter, enables optimization for specific applications.
FREMONT, CA: Industrial robots have become increasingly affordable and capable, particularly for precision assembly and high-speed picking/packing tasks. With technological advancements, each new generation of robots exhibits enhanced agility and flexibility, resembling human-like characteristics. There have been significant advancements in introducing tactile sensing capabilities to robots, allowing them to handle delicate objects and interact more safely with humans. Various techniques have been explored, including liquid metal sensors, optical-based solutions, and magnetic sensors.
These advancements aim to provide reliable and cost-effective tactile feedback, enabling industrial robots to perform a wider range of tasks effectively.
Stay ahead of the industry with exclusive feature stories on the top companies, expert insights and the latest news delivered straight to your inbox. Subscribe today.
Exploring tactile sensing techniques: Liquid metal sensors have been developed to measure the resistance of a liquid metal flowing through microfluidic channels, modulated by external forces. Although these sensors can be integrated around the robot's fingertips, they cannot measure localized 3D force. Instead, the distributed forces are mapped to a resistance change, making them suitable for specialized classification tasks following dedicated training. Optical-based solutions utilize high-end cameras to measure the deformation of elastomeric materials embedded in or covering gripping surfaces. While this technology is commercially available and used in smart multi-modal robotic grippers, it requires a substantial pixel array and demands significant communication bandwidth and power for data analysis. An alternative optical-based solution employs a quadrant of photodiode detectors instead of a full camera to address these challenges. These detectors sense the deformation of an elastic dome caused by contact forces, providing a more efficient approach. However, the power consumption remains relatively high compared to simpler alternatives.
Advancements in magnetic sensing for tactile feedback: Magnetic sensors offer a promising alternative for tactile sensing in industrial robots, offering benefits such as low power consumption and minimal computing and communication requirements. These sensors utilize an elastomeric material with an embedded magnet, similar to the approach used in camera-based systems. Using the magnetic field change resulting from the magnet displacement during elastomer deformation, 3D force sensing can be achieved. Single-output magnetometers have been demonstrated as tactile pixels, providing valuable feedback. However, these sensors are susceptible to interference from external magnetic fields, such as electric motors or other magnets.
Addressing interference with multi-pixel magnetic force sensors: Integrating multiple nearby pixels within the same integrated circuit (IC) package has been explored to overcome the limitations of single-pixel magnetic sensing. By enabling differential measurement, this configuration provides immunity to stray magnetic fields. The Melexis MLX90372 linear-displacement sensor exemplifies the gradiometric sensing principle and offers a convenient platform for demonstrating multi-pixel magnetic force sensing. The sensor's pixel readouts can be accessed directly from memory when configured in test mode. An elastomer containing an embedded disk magnet with axial magnetization is placed on top of the IC package. Applying contact force to the elastomer displaces the magnet, modulating the magnetic field pattern. The four magnetic pixels within the IC sense this displacement and enables normal and lateral force detection. Sensors measure only the magnet's displacement, while the connection between displacement and applied force requires further calibration and calculation.
More in News