Author name cluster

Chengjun Zhang

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

3 papers

2 author rows

IROS Conference 2024 Conference Paper

CASRL: Collision Avoidance with Spiking Reinforcement Learning Among Dynamic, Decision-Making Agents

Chengjun Zhang
Ka-Wa Yip
Bo Yang
Zhiyong Zhang
Mengwen Yuan
Rui Yan 0005
Huajin Tang

Developing an efficient collision avoidance policy with Spiking Reinforcement Learning for dynamic, decision-making agents remains challenging. Moreover, the implementation of energy-efficient collision avoidance is important for mobile robots that operate with limited on-board computing resources. Most existing energy-efficient methods via spiking reinforcement learning are predominately concerned with the navigational capabilities of a single agent, and are unable to handle a large, and possibly varying number of agents. To overcome these limitations, we propose a model called collision avoidance with spiking reinforcement learning (CASRL), based on proximal policy optimization algorithms. This proposed model consists of an actor with spiking neural networks (SNNs) and a critic with deep neural networks (DNNs). Our spiking reinforcement learning algorithm is advantageous to handle an arbitrary number of other agents by virtue of a spiking-gated transformer (SpikeGTr) architecture and an accumulate-to-fire (ATF) module. Extensive experimental results demonstrate that CASRL obtains a competitive success rate of navigation and exhibits higher time-efficiency for navigation in crowded scenarios compared to traditional DNN-based methods.

Details

EAAI Journal 2024 Journal Article

Complementary knowledge augmented multimodal learning method for yarn quality soft sensing

Chuqiao Xu
Linchen Xu
Shuxuan Zhao
Lianqing Yu
Chengjun Zhang

Soft sensing of yarn quality is critical for process monitoring and quality control in smart manufacturing in the textile industry. However, current methods still suffer from limitations in terms of accuracy because they only consider fixed and singular inputs related to fiber properties and process parameters. This study proposes a complementary knowledge-augmented multimodal learning method that autonomously learns complementary knowledge from the assistant modality of yarn appearance images to compensate for the deviation of the model pretrained by the primary modality of fiber and process parameters. First, a pretraining network is established to extract the feature representations of the primary modality and preliminarily output the yarn quality indices. Next, a feature fusion mechanism composed mainly of a correlation evaluation matrix and fusion gate is designed based on expert knowledge to match the complementary features as a compensation to correct the preliminary outputs. The experimental results from the workshop demonstrate that the proposed method achieves an accuracy of at least 94. 53% for the yarn appearance quality indicators. Compared with current yarn quality estimation methods, the proposed method significantly improves the accuracy by at least 2. 89%. In particular, the accuracy improved by over 6. 94% for the soft sensing of yarn neps.

Details DOI

IROS Conference 2023 Conference Paper

Spiking Reinforcement Learning with Memory Ability for Mapless Navigation

Bo Yang
Mengwen Yuan
Chengjun Zhang
Chaofei Hong
Gang Pan 0001
Huajin Tang

Our study focuses on mapless navigation in robotics, which involves navigating without an established obstacle map of the environment. Spiking Neural Networks (SNNs) have recently been applied to this task using Deep Reinforcement Learning (DRL), but face challenges in dynamic and partially observable environments, as well as inaccuracies in transmitted data. To overcome these issues, we propose a Multi-Critic DDPG with Spiking Memory (MC-DDPGSM) framework. Our approach introduces a spiking Gate Recurrent Unit layer (Spiking-GRU) to provide memory function and evaluates the state-action value with multi-critic networks. The experimental results demonstrate that our method achieves better performance (success rate, navigation distance, navigation time spent, and power consumption) in complex navigation tasks compared to the state-of-the-art approaches. Furthermore, our model can be transferred to unseen environments without the need for fine-tuning.

Details