Selected work

Representative Papers

Engineering MultiQueues: Fast Relaxed Concurrent Priority Queues

Jul 03, 2021Embedded Systems and Applications

Traditional concurrent priority queues often become performance bottlenecks under parallel workloads, limiting throughput and increasing latency. This paper proposes MultiQueues—a high-throughput, low-latency concurrent priority queue with relaxed semantics. Methodologically, it employs a multi-sharded queue architecture, triple batching (for insertions, deletions, and load balancing), and a novel wait-free locking technique to enable lock-free transformation from sequential heap structures into efficient concurrent relaxed ones. It further introduces two quantitative metrics—rank error and delay—to rigorously characterize relaxation quality. Experimental evaluation on representative workloads—including online scheduling and discrete-event simulation—demonstrates 3–10× higher throughput over state-of-the-art designs, while maintaining bounded rank error and stable, low latency. MultiQueues thus achieves superior scalability and predictability without sacrificing practicality.

6 citations3 influentialRead paper

6D Pose Estimation on Point Cloud Data through Prior Knowledge Integration: A Case Study in Autonomous Disassembly

May 30, 2025Procedia CIRP

This work addresses the challenge of 6D pose estimation for textureless, heavily occluded industrial components—such as starter motor bolts—in automated disassembly. We propose an end-to-end point cloud pose estimation framework integrating differentiable CAD model priors. Built upon the PointPillars backbone, our method innovatively incorporates rendered CAD gradient cues into the attention mechanism to enforce geometric consistency, and couples a pose graph optimization module to fuse multi-view observations and propagate global geometric constraints. Evaluated on our newly constructed AD-Pose dataset, the approach reduces mean rotation error by 42% and mean translation error by 38%, achieving real-time inference at 23 FPS. To the best of our knowledge, this is the first work to enable differentiable embedding and joint optimization of CAD geometric priors within point cloud feature matching, significantly improving robustness and accuracy for small, occluded targets in industrial settings.

3 citationsRead paper

A Cross Branch Fusion-Based Contrastive Learning Framework for Point Cloud Self-supervised Learning

Mar 18, 2024International Conference on 3D Vision

Existing self-supervised contrastive learning methods for point clouds suffer from isolated multi-branch feature encoding, where branches operate independently until the loss layer, limiting inter-branch semantic interaction. Method: We propose PoCCA—a Point Cloud Cross-Branch Attention framework—that introduces cross-branch attention at early encoder stages to enable explicit feature interaction and fusion, departing from conventional unidirectional independent encoding. PoCCA jointly optimizes point cloud augmentation strategies, a dual-branch encoder, and a sub-branch attention module, requiring no additional labels or training data. Contribution/Results: On benchmarks including ModelNet40, PoCCA achieves state-of-the-art performance in downstream classification and segmentation tasks under self-supervised settings, demonstrating that early-stage cross-branch information exchange significantly enhances representation discriminability and generalizability.

3 citationsRead paper

Safety Reinforced Model Predictive Control (SRMPC): Improving MPC with Reinforcement Learning for Motion Planning in Autonomous Driving

Sep 24, 20232023 IEEE 26th International Conference on Intelligent Transportation Systems (ITSC)

To address the limitations of conventional model predictive control (MPC) in autonomous driving motion planning—namely, restricted solution spaces due to convex approximations and the difficulty of balancing real-time performance with global optimality—this paper proposes a safety-enhanced reinforcement learning (RL) and MPC co-optimization framework. Methodologically, it incorporates an energy-function-based safety index constraint and designs state-dependent, online-updated Lagrange multipliers to embed safety requirements into both RL policy optimization and MPC solving, enabling joint safe optimization of reference trajectory generation and local control. Its key contribution is the first integration of a safety index function with an adaptive Lagrange multiplier mechanism, overcoming convex approximation constraints and enabling broader exploration of globally optimal solutions. Evaluated in highway scenarios, the approach achieves a 23.6% improvement in collision avoidance rate and an 18.4% reduction in jerk (trajectory smoothness), while maintaining millisecond-level real-time responsiveness—outperforming baseline MPC and standard safety-aware RL methods.

3 citationsRead paper

Learning conformational ensembles of proteins based on backbone geometry

Feb 19, 2025arXiv.org

Existing protein conformational sampling methods—relying either on evolutionary information or pretrained folding models—suffer from limited applicability, low efficiency, and potential biases. To address these limitations, we propose BBFlow, the first flow-matching generative model that operates exclusively on backbone geometric structure, requiring neither evolutionary sequence information nor pretrained models, and directly learns a conformational ensemble consistent with the Boltzmann distribution from scratch. BBFlow innovatively employs equilibrium backbone geometry both to condition the vector field and to define a learnable SE(3)-equivariant prior distribution, enabling robust modeling of multi-chain proteins and de novo design. Compared to state-of-the-art methods, BBFlow achieves orders-of-magnitude faster training (converging in GPU-days) and significantly accelerated inference, while maintaining competitive performance on both native protein reconstruction and de novo design benchmarks.

2 citations1 influentialRead paper

Recent publications

Latest Papers

RAFAIL: Relationship-Aware Failure Detection for Robotic Manipulation

Sep 16, 2026

RAFAIL通过检测任务相关实体间的关系异常来识别机器人操作中的执行失败，减少了对无关场景变化的敏感性，无需故障数据。

0 citationsRead paper

Calmables: Demonstrating Closed-Loop Infrared Earables for Thermal Biofeedback and Relaxation Support

Sep 16, 2026

研究通过智能戒指监测心率，并使用闭环红外耳戴设备提供局部温暖提示，以支持生物反馈和放松，帮助用户在急性激活后恢复。

0 citationsRead paper

EarStreAM: A Closed-Loop Earable System for Personalized Stress-Adaptive Meditation

Sep 16, 2026

研究提出EarStreAM系统，通过耳内生理信号监测与个性化实时干预结合的方法，实现适应性冥想以缓解压力。

0 citationsRead paper

Optimized Wrench Polytope Analysis for Real-Time Stability Control of Legged Robots in Complex Multi-Contact Configurations

Sep 15, 2026

为解决腿式机器人在复杂环境中的行走稳定性问题，提出了一种优化的力矩多边形分析算法，能够在49Hz控制频率下实时计算关节扭矩。

0 citationsRead paper

Stuffed IBLTs: Optimal Linear Multiset Sketches

Sep 15, 2026

该论文提出了一种名为Stuffed IBLT的线性草图方法，用于精确恢复原始向量。此方法在保持信息理论最优空间使用的同时，支持高效的更新和解码操作，解决了多集合协调问题。

0 citationsRead paper