The device is designed for advanced robotics applications, robotic grippers, and humanoid hands, where contact detection is crucial for handling soft, fragile, or irregularly shaped objects. The ATFCD52 supports up to 52 channels, offers sampling rates up to 800 Hz, and can detect capacitance variations of less than 1 femtofarad (fF).

Although machine vision systems allow for the location and identification of objects, robotic manipulation also needs information about contact forces.

The ATFCD52 digitizes signals from two-dimensional capacitive arrays, providing data that an external processor can use to calculate the normal and transverse forces applied during gripping. This information enables the development of systems capable of detecting when an object begins to slip and adjusting the gripping pressure to prevent it from falling or being damaged. According to Giovanni Fontana, director of Microchip's human-machine interfaces division, the ability of robots to detect and respond to physical contact remains one of the major challenges in robotics.

The executive points out that the ATFCD52 is designed to provide real-time tactile data that facilitates the development of more skillful robots and physical artificial intelligence (Physical AI) applications.

Up to 52 channels and 800 Hz sampling

The digitizer supports configurations of up to 32 transmit (TX) lines and 20 receive (RX) lines, enabling the use of dense capacitive arrays distributed across a contact surface. The device integrates 20 parallel analog input stages, capable of simultaneously acquiring signals from all receive lines. This architecture allows for sampling frequencies of up to 800 Hz, depending on the size of the sensor array.

Its main features include:

Up to 52 channels for capacitive sensors.

32 TX lines and 20 RX lines.

20 parallel analog inputs.

Sampling frequency up to 800 Hz.

Sensitivity to capacitance variations less than 1 fF.

Frequency hopping and oversampling to reduce electrical interference.

The high sensitivity facilitates the detection of small deformations in the sensors, especially during soft contacts that generate very small variations in capacitance.

Interference reduction in robotic environments

Robotic systems often incorporate servomotors, power electronics, and switched-mode controllers that can generate electrical noise and hinder the acquisition of tactile signals. To improve performance under these conditions, the ATFCD52 incorporates frequency hopping and oversampling techniques designed to attenuate interference. These features allow the use of highly sensitive capacitive sensors in environments where touch signals must be distinguished from noise generated by other electronic components.

Integration with microcontrollers, microprocessors, and FPGAs

The ATFCD52 can work alongside Microchip microcontrollers, microprocessors, and FPGAs, allowing processing power to be tailored to the requirements of each application. Developers can use it in both local control systems and architectures that demand higher computing performance. In addition to robotics, the device is geared toward augmented reality applications, gaming accessories, data gloves, touch surfaces, and wearable devices, including electronic patches.

EV-ATFCD52 development board and configuration software

Microchip offers the EV-ATFCD52 development board to facilitate the evaluation of the digitizer with different capacitive sensors, materials, and touch sensing layers.

The device also features the ATFCD52 Configurator software, which allows you to adjust operating parameters and record data in real time during testing.

More information