Human–robot interaction is becoming increasingly common to perform useful tasks in everyday life. From the human–machine communication perspective, achieving effective interaction in natural language is one challenge. To address it, natural language processing strategies have recently been used, commonly following a supervised machine learning framework. In this context, most approaches rely on the use of linguistic resources (e.g., taggers or embeddings), including training corpora. Unfortunately, such resources are scarce for some languages in specific domains, increasing the complexity of solution approaches. Motivated by these challenges, this paper explores deep learning methods for understanding natural language commands emitted to service robots that guide their movements in low-resource scenarios, defined by the use of Spanish and Nahuatl languages, for which linguistic resources are scarcely unavailable for this specific task. Particularly, we applied natural language understanding (NLU) techniques using deep neural networks and transformers-based models. As part of the research methodology, we introduced a labeled dataset of movement commands in the mentioned languages. The results show that models based on transformers work well to recognize commands (intent classification task) and their parameters (e.g., quantities and movement units) in Spanish, achieving a performance of 98.70% (accuracy) and 96.96% (F1) for the intent classification and slot-filling tasks, respectively). In Nahuatl, the best performance obtained was 93.5% (accuracy) and 88.57% (F1) in these tasks, respectively. In general, this study shows that robot movements can be guided in natural language through machine learning models using neural models and cross-lingual transfer strategies, even in low-resource scenarios.
The design and implementation of a car’s black box system using a Raspberry Pi microcomputer and an Internet of things module is presented in this research. This system was built using a Raspberry Pi microcomputer and different sensors, including a GPS, camera module, audio module, alcohol sensor module, and signals from the electronic control unit. The data were stored in both a secure digital card and in the cloud using the Waveshare SIM7600G-H 4G module. The results show that this embedded system can acquire and process video, audio, GPS data, alcohol concentration, speed, temperature, etc. Finally, a graphics user interface was developed to process the stored files. This system is similar to a black box in airplanes, which records all flight information into the black box using a specific algorithm. This makes the recorded flight data accessible to authorities when needed.
In this study, the impact of pH on the production of ZnO nanostructured thin films using chemical bath deposition was investigated for the purpose of enhancing the efficiency of solar cells. The ZnO films were directly deposited onto glass substrates at various pH levels during the synthesis process. The results indicate that the crystallinity and overall quality of the material were not affected by the pH solution, as observed through X-ray diffraction patterns. However, scanning electron microscopy revealed that surface morphology improved with increasing pH values, leading to changes in the size of the nanoflowers between pH 9 and 11 values. Furthermore, the ZnO nanostructured thin films synthesized at pH levels of 9, 10, and 11 were utilized in the fabrication of dye-sensitized solar cells. The ZnO films synthesized at pH 11 exhibited superior characteristics in short-circuit current density and open-circuit photo-voltage compared with those produced at lower pH values.
Simulations on mobility influence in optoelectronics parameters from an InGaN/GaN blue LED using the Nextnano++ software arepresented in this paper. These simulations were performed by changing the hole and electron mobility value for the material compounds according to experimental, theoretical, and doping-concentration data already reported in the literature. The power law mobility is used for the current calculation in the quantum drift-diffusion model. The results indicate the lower hole and electron leakage currents correspond to the lowest mobility values for the InGaN alloy, the greatest amount of recombination occurs in the extreme wells within the active layer of the LED and the stable emission is at 3.6 V with peak wavelength and full width at half maximum for the three mobilities. Although experimental and theoretical mobility values reach higher carrier density and recombination, the photon emission is broader and unstable. Additionally, the doping-concentration mobility results in lower wavelength shifts and narrows FWHM, making it more stable. The highest quantum efficiency achieved by doping-concentration mobility is only in the breakdown voltage (), which is the IQE value comparable to similar LEDs and is more useful for these kinds of semiconductor devices.
Written communication is pivotal for societies to develop. However, lexicon and depth of information vary greatly among texts according to their purpose. Scientific texts, diffusion of science reports, general and area-specific news are all written differently. Thus, we explore the characterization of different text categories through a nature-inspired feature known as the Hurst parameter. We contend that the Hurst exponent is useful to unveil the rhetorical structure within written documents. We collected and processed texts in five categories: scientific articles, diffusion of science reports, business news, entertainment news, and random texts. Each category contains 350 documents. We found that the median for scientific texts has the highest value of the Hurst parameter (0.575), followed by business news (0.54); the median for randomly-generated texts is 0.48, which lies in the region associated with random walks. The median value for diffusion texts is 0.49, and for entertainment texts is 0.53. However, these two categories present high dispersion. We conclude that the Hurst parameter is a measure that quantifies the structure of communication in the selected categories of texts. Application of our finding in the field of e-research is discussed.