Search for: All records

Robotic systems present unique safety challenges due to their complex integration of computational and physical processes and direct interaction with humans and environments. Traditional approaches to robot safety planning either rely on conventional methods, which struggle with the complexity of modern robotic systems, or on pure machine learning techniques, which lack formal safety guarantees. While recent advances in Large Language Models (LLMs) offer promising capabilities, pre-trained LLMs alone lack the specific domain expertise required for effective robotic safety planning. This paper introduces SafeNet, a novel neural-symbolic network architecture that enhances LLMs' safety planning capabilities through formal method-guided fine-tuning for robotic applications. Our approach integrates formal logical knowledge and reward machines into pre-trained LLMs by carefully designed fine-tuning, creating a neural-symbolic approach that combines the flexibility of neural networks with the precision of formal methods for robot trajectory generation and task planning. Experimental results demonstrate significant improvements in safe trajectory generation for robotic systems, with planning success rates increasing from 1.17% to 91.60% for the block manipulation task and from 7.23% to 90.63% for the robotic path planning task. 

In Cyber-Physical Systems (CPS), sensor data integrity is crucial since acting on malicious sensor data can cause serious consequences, given the tight coupling between cyber components and physical systems. While extensive works focus on sensor attack detection, attack diagnosis that aims to find out when the attack starts has not been well studied yet. This temporal sensor attack diagnosis problem is equally important because many recovery methods rely on the accurate determination of trustworthy historical data. To address this problem, we propose a lightweight data-driven solution to achieve real-time sensor attack diagnosis. Our novel solution consists of five modules, with the attention and diagnosis ones as the core. The attention module not only helps accurately predict future sensor measurements but also computes statistical attention scores for the diagnosis module. Based on our unique observation that the score fluctuates sharply once an attack launches, the diagnosis module determines the onset of an attack through monitoring the fluctuation. Evaluated on high-dimensional high-fidelity simulators and a testbed, our solution demonstrates robust and accurate temporal diagnosis results while incurring millisecond-level computational overhead on Raspberry Pi. 

Incremental learning is a challenging task in the field of machine learning, and it is a key step towards autonomous learning and adaptation. With the increasing attention on neuromorphic computing, there is an urgent need to investigate incremental learning techniques that can work in this paradigm to maintain energy efficiency while benefiting from flexibility and adaptability. In this paper, we present SEMINAR (sensitivity modulated importance networking and rehearsal), an incremental learning algorithm designed specifically for EMSTDP (Error Modulated Synaptic-Timing Dependent Plasticity), which performs supervised learning for multi-layer spiking neural networks (SNN) implemented on neuromorphic hardware, such as Loihi. SEMINAR uses critical synapse selection, differential learning rate and a replay buffer to enable the model to retain past knowledge while maintaining flexibility to learn new tasks. Our experimental results show that, when combined with the EMSTDP, SEMINAR outperforms different baseline incremental learning algorithms and gives more than 4% improvement on several widely used datasets such as Split-MNIST, Split-Fashion MNIST, Split-NMNIST and MSTAR. 

Spiking neural networks(SNNs) have drawn broad research interests in recent years due to their high energy efficiency and biologically-plausibility. They have proven to be competitive in many machine learning tasks. Similar to all Artificial Neural Network(ANNs) machine learning models, the SNNs rely on the assumption that the training and testing data are drawn from the same distribution. As the environment changes gradually, the input distribution will shift over time, and the performance of SNNs turns out to be brittle. To this end, we propose a unified framework that can adapt nonstationary streaming data by exploiting unlabeled intermediate domain, and fits with the in-hardware SNN learning algorithm Error-modulated STDP. Specifically, we propose a unique self training framework to generate pseudo labels to retrain the model for intermediate and target domains. In addition, we develop an online-normalization method with an auxiliary neuron to normalize the output of the hidden layers. By combining the normalization with self-training, our approach gains average classification improvements over 10% on MNIST, NMINST, and two other datasets.