Search

As the third‐generation network succeeding artificial neural networks (ANNs), spiking neural networks (SNNs) offer high robustness and low energy consumption. Inspired by biological systems, the limitations of low robustness and high‐power consumption in deep reinforcement learning (DRL) are addressed by introducing SNNs. The Brain Q‐network (BrainQN) is proposed, which replaces the neurons in the classic Deep Q‐learning (DQN) algorithm with SNN neurons. BrainQN is trained using surrogate gradient learning (SGL) and ANN‐to‐SNN conversion methods. Robustness tests with input noise reveal BrainQN's superior performance, achieving an 82.14% increase in rewards under low noise and 71.74% under high noise compared to DQN. These findings highlight BrainQN's robustness and superior performance in noisy environments, supporting its application in complex scenarios. SGL‐trained BrainQN is more robust than ANN‐to‐SNN conversion under high noise. The differences in network output correlations between noisy and original inputs, along with training algorithm distinctions, explain this phenomenon. BrainQN successfully transitioned from a simulated Pong environment to a ball‐catching robot with dynamic vision sensors (DVS). On the neuromorphic chip PAICORE, it shows significant advantages in latency and power consumption compared to Jetson Xavier NX. [ABSTRACT FROM AUTHOR]

Bacterial wilt caused by Ralstonia solanacearum is a destructive plant disease, particularly in potato (Solanum tuberosum). R. solanacearum deploys a diverse and potent arsenal of type III effectors to inhibit the plant immune system. However, the understanding of individual effectors promoting susceptibility in host plants and interfering with plant immunity responses is still limited. Here, we demonstrated that the type III effector RipV1 functioned as a novel E3 ubiquitin ligase (NEL) effector and exhibited E3 ubiquitin ligase activity in vitro. Transient expression of RipV1 suppressed plant pathogen‐associated molecular pattern (PAMP)‐triggered immunity (PTI) responses in Nicotiana benthamiana, such as the expression of PTI‐related genes and the reactive oxygen species (ROS) burst. Prolonged expression of RipV1 induced cell death in N. benthamiana leaves. Notably, mutating the conserved cysteine residue of RipV1 abolished its E3 ligase activity and its ability to suppress plant PTI responses. This study also revealed the indispensability of RipV1 for R. solanacearum's full virulence in potato. Transgenic potato plants overexpressing ripV1 but not the catalytic mutant ripV1‐C444A displayed enhanced susceptibility to R. solanacearum. RipV1 was observed to localize specifically to the plant plasma membrane, with its N‐terminus being pivotal in determining this localization. These findings showcase that RipV1 acts as a NEL effector and contributes to R. solanacearum virulence by suppressing plant PTI responses through its E3 activity. [ABSTRACT FROM AUTHOR]

The article proposes a new Kalman depth estimation and an improved swarm intelligence optimisation algorithm for adaptive tuning of servo gain for image-based visual servo control. First, a Kalman depth estimation model is established from the principle of image-based visual servoing, and two state equations are designed for depth estimation based on the number of state quantities. Second, the improved sparrow search algorithm is proposed to tune the servo gain adaptively to improve the convergence speed and stability. To verify the effectiveness of the proposed method, the conventional image-based visual servoing and conventional Kalman estimation are reproduced and compared with the proposed method, and the simulation is completed on the Simulink simulation platform for verification. Finally, the experiments are completed in the robotic arm experimental platform. Both the simulation and experimental results show the effectiveness of the proposed method, which reduces the redundancy of the camera and shortens the convergence time. [ABSTRACT FROM AUTHOR]

In this article, KNNLT-BNT lead-free ceramics were prepared by the solid solution method to regulate the grain size and electrical properties by Mn-doped. The doped concentrations were 0.3, 0.4, and 0.5%. The surface structure of the ceramic was characterized, and the Mn-doped can improve the ceramic compactness. The internal structure of the ceramic was analyzed, with the increase of Mn-doped concentration, the (101) peak shifted to a lower angle, and the lattice is distorted. Finally, the electrical properties of the ceramic were characterized by the optical band gap and dielectric constant. Analysis results show that when the Mn-doped concentration is 0.5%, the electrical performance is the best, the grain size follows a normal distribution, and the band gap width is the narrotest 2.693 eV. [ABSTRACT FROM AUTHOR]