Robotic computing systems play an important role in enabling intelligent robotic tasks through intelligent algorithms and supporting hardware. In recent years, the evolution of robotic algorithms indicates a roadmap from traditional robotics to hierarchical and end-to-end models. This algorithmic advancement poses a critical challenge in achieving balanced system-wide performance. Therefore, algorithm-hardware co-design has emerged as the primary methodology, which analyzes algorithm behaviors on hardware to identify common computational properties. These properties can motivate algorithm optimization to reduce computational complexity and hardware innovation from architecture to circuit for high performance and high energy efficiency. We then reviewed recent works on robotic and embodied AI algorithms and computing hardware to demonstrate this algorithm-hardware co-design methodology. In the end, we discuss future research opportunities by answering two questions: (1) how to adapt the computing platforms to the rapid evolution of embodied AI algorithms, and (2) how to transform the potential of emerging hardware innovations into end-to-end inference improvements.

The high-power microwave (HPM) effect heats solar cells, which is an important component of a satellite. This creates a serious reliability problem and affects the normal operation of a satellite. In this paper, the different HPM response characteristics of two kinds of solar cells are comparatively researched by simulation. The results show that there are similarities and differences in hot spot distribution and damage mechanisms between both kinds of solar cell, which are related to the amplitude of HPM. In addition, the duty cycle of repetition frequency contributes more to the temperature accumulation of the solar cells than the carrier frequency. These results will help future research of damage assessment technology, reliability enhancement and the selection of materials for solar cells.

A wide wavelength tuning range and single-mode hybrid cavity laser consists of a square Whispering-Gallery (WG) microcavity and a Fabry–Pérot (FP) was introduced and demonstrated. A wavelength tuning range over 12.5 nm from 1760.87 to 1773.39 nm which was single-mode emitting was obtained with the side-mode suppression ratio over 30 dB. The hybrid cavity laser does not need grating etching and special epitaxial structure, which reduces the fabrication difficulty and cost, and shows the potential for gas sensing with absorption lines in this range.

The electron mobility anisotropy in (Al,Ga)Sb/InAs two-dimensional electron gases with different surface morphology has been investigated. Large electron mobility anisotropy is found for the sample with anisotropic morphology, which is mainly induced by the threading dislocations in the InAs layer. For the samples with isotropic morphology, the electron mobility is also anisotropic and could be attributed to the piezoelectric scattering. At low temperature (below transition temperature), the piezoelectric scattering is enhanced with the increase of temperature, leading to the increase of electron mobility anisotropy. At high temperature (above transition temperature), the phonon scattering becomes dominant. Because the phonon scattering is isotropic, the electron mobility anisotropy in all the samples would be reduced. Our results provide useful information for the comprehensive understanding of electron mobility anisotropy in the (Al,Ga)Sb/InAs system.

Germanene nanostrips (GeNSs) have garnered significant attention in modern semiconductor technology due to their exceptional physical characteristics, positioning them as promising candidates for a wide range of applications. GeNSs exhibit a two-dimensional (buckled) honeycomb-like lattice, which is similar to germanene but with controllable bandgaps. The modeling of GeNSs is essential for developing appropriate synthesis methods as it enables understanding and controlling the growth process of these systems. Indeed, one can adjust the strip width, which in turn can tune the bandgap and plasmonic response of the material to meet specific device requirements. In this study, the objective is to investigate the electronic behavior and THz plasmon features of GeNSs (≥100 nm wide). A semi-analytical model based on the charge-carrier velocity of freestanding germanene is utilized for this purpose. The charge-carrier velocity of freestanding germanene is determined through the GW approximation (${v}_{\mathrm{F}}=0.702 \times {10}^{6} $ m·s⁻¹). Within the width range of 100 to 500 nm, GeNSs exhibit narrow bandgaps, typically measuring only a few meV. Specifically, upon analysis, it was found that the bandgaps of the investigated GeNSs ranged between 29 and 6 meV. As well, these nanostrips exhibit $ \sqrt{q} $-like plasmon dispersions, with their connected plasmonic frequency (≤30 THz) capable of being manipulated by varying parameters such as strip width, excitation plasmon angle, and sample quality. These manipulations can lead to frequency variations, either increasing or decreasing, as well as shifts towards larger momentum values. The outcomes of our study serve as a foundational motivation for future experiments, and further confirmation is needed to validate the reported results.

A new semi-insulation structure in which one isolated island is connected to the substrate was proposed. Based on this semi-insulation structure, an advanced BCD technology which can integrate a vertical device without extra internal interconnection structure was presented. The manufacturing of the new semi-insulation structure employed multi-epitaxy and selectively multi-doping. Isolated islands are insulated with the substrate by reverse-biased PN junctions. Adjacent isolated islands are insulated by isolation wall or deep dielectric trenches. The proposed semi-insulation structure and devices fixed in it were simulated through two-dimensional numerical computer simulators. Based on the new BCD technology, a smart power integrated circuit was designed and fabricated. The simulated and tested results of Vertical DMOS, MOSFETs, BJTs, resistors and diodes indicated that the proposed semi-insulation structure is reasonable and the advanced BCD technology is validated.

Due to the quantum confinement effect, atomically thin two-dimensional (2D) monolayer materials possess distinct characteristics from their corresponding bulk materials, which have received wide attention from science and industry. Among all the 2D materials, elemental 2D materials with the simplest components are most striking. As an emerging group-VIA elemental 2D monolayer material, tellurene exhibits many exciting fundamental properties, such as chemical and mechanical stabilities, bandgap and high carrier mobilities compared to phosphorene, graphene and MoS₂, respectively. Besides, in further exploration, it was found that tellurene or tellurene-based device presents excellent thermoelectric properties, piezoelectric properties, quantum Hall effects, and superb optical properties especially nonlinear optics characteristics, etc. The properties of tellurene can be modulated by virtue of strain, defects, edges, and heterojunction effects. In view of so many unique properties, it has drawn significant interest since tellurene was predicted and fabricated successfully in 2017. In this paper, we review the 2D tellurene allotropes, experimental preparation, excellent properties, performance modulation and future development.

In this work, we studied the persistent photoconductivity (PPC) spectra in single HgTe/CdHgTe quantum wells with different growth parameters and different types of dark conductivity. The studies were performed in a wide radiation quantum energy range of 0.62–3.1 eV both at T = 4.2 K and at T = 77 K. Common features of the PPC spectra for all structures were revealed, and their relation to the presence of a CdTe cap layer in all structures and the appropriate cadmium fraction in the CdHgTe barrier layers was shown. One of the features was associated with the presence of a deep level in the CdTe layer. In addition, the oscillatory behavior of the PPC spectra in the region from 0.8–1.1 eV to 1.2–1.5 eV was observed. It is associated with the cascade emission of longitudinal optical phonons in CdHgTe barrier.

Effective detection of methamphetamine (Met) requires a fast, sensitive, and cheap testing assay. However, commercially available methods require expensive instruments and highly trained operators, which are time-consuming and labor-intensive. Herein, an antibody-modified graphene transistor assay is developed for sensitive and minute-level detection of Met in complex environments. The anti-Met probe captured charged targets within 120 s, leading to a p-doping effect near the graphene channel. The limit of detection reaches 50 aM (5.0 × 10⁻¹⁷ M) Met in solution. The graphene transistor would be a valuable tool for Met detection effective prevention of drug abuse.