Large Language Models (LLMs) have gained popularity in task planning for long-horizon manipulation tasks. To enhance the validity of LLM-generated plans, visual demonstrations and online videos have been widely employed to guide the planning process. However, for manipulation tasks involving subtle movements but rich contact interactions, visual perception alone may be insufficient for the LLM to fully interpret the demonstration. Additionally, visual data provides limited information on force-related parameters and conditions, which are crucial for effective execution on real robots. In this paper, we introduce LEMMo-Plan, an in-context learning framework that incorporates tactile and force-torque information from human demonstrations to enhance LLMs' ability to generate plans for new task scenarios. We propose a bootstrapped reasoning pipeline that sequentially integrates each modality into a comprehensive task plan. This task plan is then used as a reference for planning in new task configurations. Real-world experiments on two different sequential manipulation tasks demonstrate the effectiveness of our framework in improving LLMs' understanding of multi-modal demonstrations and enhancing the overall planning performance. More materials are available on our project website: lemmo-plan.github.io/LEMMo-Plan/.

The battery-free and crystal-less nodes require energy-efficient radio, leading to the design of the ultra-low-power crystal-less receiver. The paper proposes a crystal-less receiver classification method for state-of-the-art works and presents a sub-1mW Class-Ab crystal-less receiver architecture with a carrier-auxiliary intermediate frequency (IF) negative feedback local oscillator (LO) generator and approximate low-IF mode. The proposed LO generator and receiver are implemented using 55nm CMOS technology with a total power of 492.4µW. Without any crystals, the LO frequency based on the bulk acoustic wave (BAW) oscillator maintains a ±6ppm error range from 0°C to 70°C. After locking the LO calibration loop, the phase noise at the 1MHz offset is better than -137dBc/Hz and long-term frequency drift is less than ±0.06ppm in two seconds.

This paper introduces a UAV-based multispectral model, YOLO-SPAD, for rapid, non-destructive estimation of relative chlorophyll content of tea leaves. This is crucial for predicting growth conditions, implementing precise irrigation and fertilization, and increasing tea yields. The model leverages the spectral separation of the tea tree canopy and the YOLOv8 neural network architecture for image segmentation. Multispectral imagery of the standardized tea plantation was captured by a UAV equipped with a five-channel camera, while corresponding SPAD values were measured using a SPAD-502Plus device. Based on the YOLOv8 model, it is proposed to incorporate a Spectral Fusion module to better adapt to multispectral characteristics. Four deep learning models for image segmentation were further compared. The improved YOLOv8 model achieved a segmentation mIoU of 91.61% for the tea tree canopy, outperforming the YOLOv5 (90.65%) and DeepLabv3+ (86.14%), which were improved using the same method. Five different band combinations, 43 vegetation indexes, and 40 texture features were analyzed to construct mapping transformation pairs for reflectance in the segmented area. This was applied to the SPAD prediction head to implement YOLO-SPAD. The YOLO-SPAD model was used to predict tea canopy SPAD using UAV multispectral imagery with a coefficient of determination (R²) of 0.85 and a low error (RMSE=2.14, MAE=1.79), providing accurate and stable predictions. This model supports dynamic monitoring of tea tree growth via UAV remote sensing, aiding crop nutrition improvement and precision agriculture implementation.

In the field of Learning from Demonstration (LfD), Dynamical Systems (DSs) have gained significant attention due to their ability to generate real-time motions and reach predefined targets. However, the conventional convergence-centric behavior exhibited by DSs may fall short in safety-critical tasks, specifically, those requiring precise replication of demonstrated trajectories or strict adherence to constrained regions even in the presence of perturbations or human intervention. Moreover, existing DS research often assumes demonstrations solely in Euclidean space, overlooking the crucial aspect of orientation in various applications. To alleviate these shortcomings, we present an innovative approach geared toward ensuring the safe execution of learned orientation skills within constrained regions surrounding a reference trajectory. This involves learning a stable DS on SO(3), extracting time-varying conic constraints from the variability observed in expert demonstrations, and bounding the evolution of the DS with Conic Control Barrier Function (CCBF) to fulfill the constraints. We validated our approach through extensive evaluation in simulation and showcased its effectiveness for a cutting skill in the context of assisted teleoperation.

In resource exchange networks, nodes or agents may cooperate or compete with each other to maximize their own profits. In competitive networks, they may determine their exchange strategy selfishly, which makes it a challenge problem to design fair and efficient cooperation strategy. A popular resource exchanging mechanism called Max-Min Fairness-Based Coalitional (MMFC) is proposed in [1] to solve the problem in competitive networks, however it is still an open problem whether the mechanism is trustful if an agent lies about its resource information. In this paper, we demonstrate the trustfulness of the MMFC mechanism; combining theoretical analyses and numerical examples, we show that an agent cannot gain more benefit by misreporting its resource information.

This paper presents a novel simultaneous multithreading (SMT) VLIW DSP architecture with dynamic dispatch mechanism to address the challenge of the underutilization of computing resources in the non-unit assumed latency (NUAL) VLIW DSPs. The SMT technology exploits the unused instruction slots by converting the thread-level parallelism to the instruction-level parallelism, improving the efficiency. With the specifically designed registers for eliminating the horizontal dependencies among the execution-packet, the NUAL VLIW DSP architecture supports issuing any subset of instructions of the execution-packet based on the availability of the corresponding functional units. With the dynamic dispatch mechanism, the DSP issues instructions to functional unit at run-time rather than at compile-time, such that the issue conflicts among multiple threads are reduced significantly. The new VLIW DSP architecture is implemented and evaluated, and the results show that the architecture can effectively increase the processor throughput, hide the cache miss latencies, and improve the performance on digital signal processing.