How can robots learn dexterous grasping skills efficiently and apply them adaptively based on user instructions? This work tackles two key challenges: efficient skill acquisition from limited human demonstrations and context-driven skill selection. We introduce AdaDexGrasp, a framework that learns a library of grasping skills from a single human demonstration per skill and selects the most suitable one using a vision-language model (VLM). To improve sample efficiency, we propose a trajectory following reward that guides reinforcement learning (RL) toward states close to a human demonstration while allowing flexibility in exploration. To learn beyond the single demonstration, we employ curriculum learning, progressively increasing object pose variations to enhance robustness. At deployment, a VLM retrieves the appropriate skill based on user instructions, bridging low-level learned skills with high-level intent. We evaluate AdaDexGrasp in both simulation and real-world settings, showing that our approach significantly improves RL efficiency and enables learning human-like grasp strategies across varied object configurations. Finally, we demonstrate zero-shot transfer of our learned policies to a real-world PSYONIC Ability Hand, with a 90% success rate across objects, significantly outperforming the baseline.

This article introduces the experiment design for Moon imaging based on Sanya incoherent scatter radar (SYISR) and algorithm research in data processing. The peak power of SYISR is 2 MW. The transmitted frequency used for Moon imaging experiments is 430 MHz. We conducted Moon imaging experiments using two types of waveforms, 13-bit Barker code, and linear frequency modulation (LFM) chirp. Considering both resolution and signal-to-noise ratio (SNR), the use of an LFM chirp with a bandwidth of 0.3 MHz and a pulsewidth of 2 ms can give higher SNR and resolution for Moon imaging using SYISR. Several key techniques were applied in the experiment design and data processing: 1) for the reliability of the imaging algorithm, the range-Doppler imaging algorithm commonly used in synthetic aperture imaging was applied; 2) to avoid the sidelobe effect of the 13-bit Barker code matched filter, a sidelobe-free filter was used; and 3) to mitigate the problem of “north–south ambiguity,” mosaic imaging of the Doppler northern and southern hemispheres of the nearside of the Moon was adopted. Two types of imaging results are obtained: mosaic images of the northern and southern hemispheres of the Moon and local regional images. The results demonstrate the feasibility and reliability of Moon imaging based on SYISR, which enables potential further lunar geology investigations in the future.

Incoherent scatter radar (ISR), as one of the most powerful ionospheric detection methods on the ground, can obtain a variety of important ionospheric parameters with high spatiotemporal resolution over the entire ionosphere. A high-power large-aperture phased array (PA) ISR, named Sanya ISR (SYISR), is under construction at Sanya (18.3°N, 109.6°E), China. Two remote receive-only arrays of SYISR will be built in the future to form a tristatic ISR system, which will be the only multistatic ISR in a low-latitude region. Estimating the signal-to-noise ratio (SNR) is the key task for analyzing the detection accuracy and evaluating the radar performance. In this article, the SNR of the SYISR tristatic system is simulated for the first time using a suitable PA bistatic radar equation. The parameters of the designed SYISR tristatic system are used in the simulation. Subsequently, the SNR spatial variability of the bistatic PA radar and bistatic equivalent parabolic dish (EPD) radar is analyzed and compared. The results show that the radar system constants, effective scattering volume, and spatial variable terms are the three different sources of factors affecting the SNR. They jointly determine the spatial distribution characteristics of the SNR. Moreover, the PA SNR is significantly different from the EPD SNR numerically and morphologically. The PA SNR is weaker than the EPD SNR in the whole observational space. In the horizontal direction, the PA SNR has a single-peak structure, while the EPD SNR has a double-peak structure. The root cause of these differences is that the beamwidth of a PA changes with the beam directions, while that of the EPD is fixed. This research provides a theoretical basis for the design of tristatic PA ISR in hardware optimization and error analysis, which will be shown in another article.

This paper is concerned with construction of efficiently encodable nonbinary quasi-cyclic LDPC codes based on finite fields. Four classes of nonbinary quasi-cyclic LDPC codes are constructed. Experimental results show that codes constructed perform well with iterative decoding using a fast Fourier transform based q-ary sum-product algorithm and they achieve significant coding gains over Reed-Solomon codes of the same lengths and rates decoded with either algebraic hard- decision Berlekamp-Massey algorithm or algebraic soft-decision Kotter-Vardy algorithm.

This paper presents five methods for constructing nonbinary LDPC codes based on finite geometries. These methods result in five classes of nonbinary LDPC codes, one class of cyclic LDPC codes, three classes of quasi-cyclic LDPC codes and one class of structured regular LDPC codes. Experimental results show that constructed codes in these classes decoded with iterative decoding based on belief propagation perform very well over the AWGN channel and they achieve significant coding gains over Reed-Solomon codes of the same lengths and rates with either algebraic hard-decision decoding or Kotter-Vardy algebraic soft-decision decoding at the expense of a larger decoding computational complexity.

In the late 1950s and early 1960s, finite fields were successfully used to construct linear block codes, especially cyclic codes, with large minimum distances for hard-decision algebraic decoding, such as Bose-Chaudhuri-Hocquenghem (BCH) and Reed-Solomon (RS) codes. This paper shows that finite fields can also be successfully used to construct algebraic low-density parity-check (LDPC) codes for iterative soft-decision decoding. Methods of construction are presented. LDPC codes constructed by these methods are quasi-cyclic (QC) and they perform very well over the additive white Gaussian noise (AWGN), binary random, and burst erasure channels with iterative decoding in terms of bit-error probability, block-error probability, error-floor, and rate of decoding convergence, collectively. Particularly, they have low error floors. Since the codes are QC, they can be encoded using simple shift registers with linear complexity.

This paper presents a general and three specific algebraic methods for constructing efficiently encodable non-binary quasi-cyclic LDPC codes. Three classes of quasi-cyclic LDPC codes over nonbinary finite fields are constructed. codes constructed perform very well over the AWGN channel with iterative decoding and achieve large coding gains over the Reed-Solomon codes of the same parameters. Nonbinary LDPC codes may be used to replace Reed-Solomon codes in some communication environments or storage systems for combating mixed types of noises and interferences.

This paper presents three algebraic methods for constructing nonbinary quasi-cyclic (QC)-LDPC codes. Three classes of efficiently encodable QC-LDPC codes over nonbinary finite fields are constructed. Experimental results show that constructed codes decoded with iterative decoding perform well over the AWGN channel and they achieve significant coding gains over Reed-Solomon (RS) codes of the same lengths and rates decoded with algebraic decoding

Quasi-cyclic (QC) low-density parity-check (LDPC) codes form an important subclass of LDPC codes. These codes have encoding advantage over other types of LDPC codes. This paper addresses the issue of efficient encoding of QC-LDPC codes. Two methods are presented to find the generator matrices of QC-LDPC codes in systematic-circulant (SC) form from their parity-check matrices, given in circulant form. Based on the SC form of the generator matrix of a QC-LDPC code, various types of encoding circuits using simple shift registers are devised. It is shown that the encoding complexity of a QC-LDPC code is linearly proportional to the number of parity bits of the code for serial encoding, and to the length of the code for high-speed parallel encoding.

This paper is concerned with construction of quasi-cyclic (QC) low-density parity-check (LDPC) codes for three different types of channels: the additive white Gaussian noise, the binary random erasure, and the binary burst erasure channels. Two algebraic methods for systematic construction of QC-LDPC codes are presented. Codes constructed perform well over all three types of channels

This paper presents methods for efficient encoding of quasi-cyclic LDPC codes. Based on these methods, encoding of quasi-cyclic LDPC codes can be implemented using simple shift-registers with complexity linearly proportional to the number of parity-check bits of a code for serial encoding and to the length of a code for parallel encoding. Various encoding circuits are devised and they provide a range of trade-offs between encoding complexity and speed.

This paper presents three algebraic methods for constructing q-ary LDPC codes. The first method gives a class of dispersed Reed-Solomon codes as LDPC codes. The second method gives a class of q-ary quasi-cyclic LDPC codes. The third method gives two classes of q-ary finite geometry LDPC codes. Codes constructed by these methods perform very well with iterative decoding, even for short codes.

The development of a system for automatically locating and tracking a human in the vicinity of a robot is described. The system consists of multiple passive infrared (PIR) sensors, two color cameras, a pair of microwave sensors and a pair of PCs for data collection, signal processing and data fusion. The cameras are treated as individual sensors rather than a stereo pair to minimize the affect of occlusion by the robot. The area around the robot is subdivided into an occupancy grid with 0.5m by 0.5m cells. A data fusion algorithm, based on Dempster-Shafer evidence theory, is used to estimate the probability of human occupancy for each cell. This information is used to estimate the human’s location. A novel concept termed a “protective cell” is introduced to further increase the human’s safety in the presence of sensor uncertainty. Experimental results are included demonstrating the system’s effectiveness.

This paper presents two algebraic methods for constructing efficiently encodable quasi-cyclic (QC) LDPC codes that perform well on both the AWGN and binary erasure channels with iterative decoding in terms of bit-error performance, block error performance and error-floor, collectively. The constructions are based on the cyclic subgroups of the multiplicative groups of finite fields and affine mappings

This paper presents a method for constructing structured LDPC codes based on finite fields and affine mappings. The Tanner graphs of these codes have girth at least six. Experimental results show that codes constructed based on this method with iterative decoding perform well on both the AWGN and binary erasure channels. Furthermore, codes constructed based on prime fields are quasi-cyclic.

Efficient Encoding of Quasi-Cyclic Low-Density Parity-Check Codes Quasi-cyclic (QC) low-density parity-check (LDPC) codes form an important subclass of LDPC codes. These codes have encoding advantage over other types of LDPC codes. This paper addresses the issue of efficient encoding of QC-LDPC codes. Two methods are presented to find the generator matrices of QC-LDPC codes in systematic-circulant form from their parity-check matrices given in circulant form. Based on the systematic-circulation form of the generator matrix of a QC-LDPC code, various types of encoding circuits using simple shift registers are devised. It is shown that the encoding complexity of a QC-LDPC code is linearly proportional to the number of parity bits of the code for serial encoding, and to the length of the code for high-speed parallel encoding.

This paper presents a superposition method for constructing low-density parity-check (LDPC) codes. Several classes of structured LDPC codes are constructed. Codes in these classes perform well with iterative decoding, and their Tanner graphs have girth at least six.

We propose a novel iterative decoding algorithm based on partial combination of the parity check matrix for iterative soft decoding of Reed-Solomon product codes. It uses only the simple syndrome equations whose solutions are confined to parts of the columns of the parity check matrix. Compared to other algorithms, the proposed algorithm has lower complexity while offering better performance, which is demonstrated by simulations.