SRIS Lab
Smart, Reliable, and Interpretable Systems — we fuse logic, physics, and learning to build trustworthy AI for intelligent manufacturing and critical equipment.
Our unified methodology—Signal Temporal Logic (STL), causal structure, and physics-guided priors—drives interpretable diagnosis, STL-driven safe autonomy, verifiable machine learning, and fleet-level health management.
Verifiable
Industrial-Grade
Interpretable
South China University of Technology • Guangzhou International Campus
Neuro-Symbolic Models
STL-regularized pretraining, causal KGs, and rule distillation for audit-ready industrial AI.
Causal-Logic PHM
Physics-guided features + causal graphs + STL abstractions to reconstruct propagation chains.
Safe Autonomy
STL contracts, reference governors, and residual RL for dual-arm & mobile manipulation.
SRIS is led by Dr. Gang Chen. We study machine learning, formal methods, control, and signal processing for cyber-physical system monitoring and control. Our deliverables include interpretable fault diagnosis algorithms, contract-based controllers, verifiable ML toolchains, and fleet-level PHM (Prognostics & Health Management).
Recent News
- Aug 15, 2025 Our paper Instantaneous frequency extraction of overlapped multicomponent signals with an adaptive constraints approach has been accepted by Engineering Research Express.
- Aug 4, 2025 Decomposition-Based MPC for Uncertain Systems with Nested STL Specifications accepted by IEEE Control Systems Letters.
- Jul 12, 2025 Event driven causal knowledge graphs for bottleneck root cause analysis in manufacturing systems accepted by Expert Systems with Applications.
- Jun 27, 2025 Congratulations to our graduates: Li Cai, Li Du, Zeng Ziyue.
- Jun 19, 2025 A dual spatial temporal neural network for bottleneck prediction in manufacturing systems accepted by EAAI.
- Jan 14, 2025 New grant from State Key Lab of Performance Monitoring and Protecting of Rail Transit Infrastructure on gearbox fault diagnosis.
- Jan 9, 2025 Temporal Logic Inference for Interpretable Fault Diagnosis of Bearings via Sparse and Structured Neural Attention accepted by ISA Transactions.
- Dec 22, 2024 Multi-source data fusion monitoring system for super-elevation in LPBF accepted by Optics and Lasers in Engineering.
- Oct 14, 2024 Fault tracing in multistage gearbox systems accepted by Measurement.
- Sep 25, 2024 In-situ quality inspection for LPBF with depth graph network accepted by MSSP.
- Jul 12, 2024 VNCCD: Gearbox diagnosis under nonstationary conditions accepted by MSSP.
- Jun 29, 2024 Survey Enhancing Reliability through Interpretability… accepted by IEEE Access.
- May 13, 2024 Fast Pareto Set Approximation… accepted by Swarm and Evolutionary Computation.