Self-Powered Pressure–Temperature Bimodal Sensing Based on the Piezo-Pyroelectric Effect for Robotic Perception

Xiang Yu, Yun Ji, Xinyi Shen, Xiaoyun LeSensors 2024 · DOI

TL;DR

Researchers have developed a novel self-powered sensor using PMN-PT single crystals that can simultaneously detect both pressure and temperature. This bimodal sensor leverages the distinct response speeds of its piezoelectric and pyroelectric effects to accurately classify complex stimuli, paving the way for more sophisticated robotic perception and manipulation.

What Problem Does This Paper Address?

Robots need to understand their environment comprehensively to interact safely and effectively. While many sensors can detect individual physical stimuli like pressure or temperature, integrating multiple sensors adds complexity and power consumption. The challenge is to create compact, low-power, multifunctional sensors that can simultaneously capture diverse environmental information. Existing approaches often rely on combining different materials, which can be cumbersome. This paper addresses the need for a single material-based sensor capable of bimodal pressure and temperature sensing, specifically for applications in robotic perception where distinguishing between different types of touch and thermal cues is crucial.

Methods

The core of this work is the use of ferroelectric Pb(Mg<sub>1/3</sub>Nb<sub>2/3</sub>)O<sub>3</sub>-PbTiO<sub>3</sub> (PMN-PT) single crystals. These materials exhibit both significant piezoelectric (pressure sensing) and pyroelectric (temperature sensing) properties. The researchers fabricated a simple three-layer sensor by sandwiching a PMN-PT thin disk between silver electrodes.

The key innovation lies in exploiting the difference in response speeds between the piezoelectric and pyroelectric effects. The piezoelectric response to pressure is very fast (on the order of milliseconds), while the pyroelectric response to temperature changes is slower (on the order of seconds). This temporal difference allows the sensor to distinguish between pressure-induced signals and temperature-induced signals, even when they occur simultaneously.

For robotic applications, the sensor was integrated into a robotic hand. A thermoelectric module was used to apply mixed force-thermal stimuli. The generated electrical signals were then fed into a fully connected Artificial Neural Network (ANN) model (specifically, a multilayer perceptron) for stimuli classification.

Datasets

The paper does not explicitly mention the creation of new benchmark datasets. Instead, it focuses on generating data from the developed sensor under controlled stimuli. The performance of the ANN classifier was evaluated using datasets created by applying six different types of stimuli (combinations of hot/cold and high/low pressure) 100 times each. These datasets were split into training (80%) and validation (20%) sets.

Key Insights and Ideas

  • Exploiting Temporal Differences: The most significant insight is the clever use of the inherent speed difference between piezoelectric and pyroelectric effects to decouple and simultaneously measure pressure and temperature. This temporal multiplexing is a powerful strategy for creating multifunctional sensors.
  • Single-Material Multifunctionality: Demonstrating that a single ferroelectric material like PMN-PT can achieve high performance in both pressure and temperature sensing without needing to combine multiple materials is a major step towards miniaturization and integration.
  • Robotic Perception Enhancement: The successful integration of the sensor into a robotic hand and its ability to classify complex stimuli (e.g., "hot and high-pressure") highlights its direct applicability to improving robotic dexterity and interaction capabilities.
  • Self-Powered Operation: The sensor operates without an external power source, relying solely on the generated electrical signals from mechanical and thermal stimuli. This is crucial for low-power, long-term robotic applications.

Key Results

  • High Sensitivity:
    • Pressure sensing sensitivity: 28.4 nA kPa<sup>-1</sup>
    • Temperature sensing sensitivity: 94.66 nC °C<sup>-1</sup>
  • Fast Response:
    • Pressure response time: 46 ms
    • Temperature response time: 1.98 s
  • Stimuli Classification Accuracy: The ANN model achieved a remarkable recognition accuracy of 99.16% for classifying six different types of mixed-force thermal stimuli.
  • Robustness: The sensor demonstrated excellent long-term stability and reproducibility for both pressure and temperature sensing.

Limitations and Future Work

  • Response Time Trade-off: While the piezoelectric response is very fast, the pyroelectric response is slower. This limits the speed at which rapid temperature changes can be accurately captured alongside pressure.
  • Complexity of Mixed Stimuli: While the system can classify distinct stimuli, accurately quantifying both pressure and temperature simultaneously in real-time from a single complex signal might still require advanced signal processing.
  • Future Work: The authors suggest that this work paves the way for developing more advanced multifunctional self-powered sensors for low-consumption artificial perception. Future research could focus on further optimizing the material properties, exploring different device architectures, and expanding the range of detectable stimuli. The integration into more complex robotic systems and real-world scenarios would also be a logical next step.

Summary updated

Abstract

Multifunctional sensors have played a crucial role in constructing high-integration electronic networks. Most of the current multifunctional sensors rely on multiple materials to simultaneously detect different physical stimuli. Here, we demonstrate the large piezo-pyroelectric effect in ferroelectric Pb(Mg<sub>1/3</sub>Nb<sub>2/3</sub>)O<sub>3</sub>-PbTiO<sub>3</sub> (PMN-PT) single crystals for simultaneous pressure and temperature sensing. The outstanding piezoelectric and pyroelectric properties of PMN-PT result in rapid response speed and high sensitivity, with values of 46 ms and 28.4 nA kPa<sup>-1</sup> for pressure sensing, and 1.98 s and 94.66 nC °C<sup>-1</sup> for temperature detection, respectively. By leveraging the distinct differences in the response speed of piezoelectric and pyroelectric responses, the piezo-pyroelectric effect of PMN-PT can effectively detect pressure and temperature from mixed-force thermal stimuli, which enables a robotic hand for stimuli classification. With appealing multifunctionality, fast speed, high sensitivity, and compact structure, the proposed self-powered bimodal sensor therefore holds significant potential for high-performance artificial perception.
BibTeX
@inproceedings{yu2024self,
  author = {Xiang Yu and Yun Ji and Xinyi Shen and Xiaoyun Le},
  title = {{Self-Powered Pressure–Temperature Bimodal Sensing Based on the Piezo-Pyroelectric Effect for Robotic Perception}},
  booktitle = {Sensors},
  year = {2024},
  doi = {10.3390/s24092773},
}
Figure