Yubo Lian

Yubo Lian

Auto-generated profile, based on publicly available information

Yubo Lian

BYD Auto Industry Company Ltd., Shenzhen, China

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
11 June 2024 in IEEE Transactions on Transportation Electrification
Verified
Journal Article
Closed access
YCYingfeng Cai
Yingfeng Cai
YLYuxing Li
Yuxing Li
YLYubo Lian
Yubo Lian
LCLong Chen
Long Chen
YZYilin Zhong
Yilin Zhong
CYChaochun Yuan
Chaochun Yuan
Four-wheel driving/four-wheel steering vehicle has the advantages of high-speed stability, steering response, and safety, making them perform excellently under various driving conditions. However, it also faces challenges due to complex dynamic couplings, multiple control inputs, and computational burden. To address these problems, this paper proposes a multi-agent cooperative control architecture based on Cooperative Distributed Model Predictive Control (Co-DMPC). The paper explains the coupling mechanisms between agents by establishing distributed state equations and reduces the control inputs of each controller. It defines predicted trajectories, assumed trajectories, and optimal trajectories for states and control inputs, laying the foundation for the interaction of coupled information between agents. The impact of different agents on global performance indicators is measured by solving the cost coupling optimization problem. Furthermore, the iterative solution approach and the termination conditions address the balance between solution accuracy and computational efficiency. The Lyapunov function direct method is used to prove the asymptotic stability of both local and global systems. This paper quantitatively analyzes the influence of weight coefficients on control effects and presents an adaptive weight selection method based on road adhesion coefficient. Compared with centralized control, the proposed method has excellent tracking accuracy, response speed, and stability. Furthermore, the hardware-in-loop (HIL) test also demonstrate the performance of the proposed controller.

9 December 2021 in E3S Web of Conferences
Verified
Conference Paper
Open access
PDF file
HWHuatao Wen
Huatao Wen
SZSongzhao Zhang
Songzhao Zhang
YLYubo Lian
Yubo Lian
ZZZhiguang Zhao
Zhiguang Zhao
WWWeibo Wang
Weibo Wang
SDSanbao Dong
Sanbao Dong
In this paper, graphene oxide (GO) was modified with alkyl amidopropyl diethanolamine to obtain a nano pour point depressant (GO-PPD), which was used to improve the flowability of the waxy oil extracted from Changqing Oilfield, China. Fourier transform infrared (FTIR), differential scanning calorimetry (DSC), polarized optical microscopy (POM) and viscometer were employed to evaluate the performance of the GOPPD. The results showed that compared with traditional pour point depressant (PPD), the GO-PPD exhibited higher performance in promoting the flowability of waxy crude oil. With the presence of 500 mg/kg GO-PPD in the waxy crude oil, the pour point of which could be reduced by 5.5 °C. Also, with the presence of 500 mg/kg GO-PPD, the viscosity reduction rate of the waxy crude oil can reach up to 52% at 30 °C. Through the observation via polarized microscopy, we have also found that with the introduction of GO-PPD in the crude oil, the formation of the wax crystals can be greatly retarded. This confirmed that the graphene oxide derivates could also be served as PPD, which facilitates the flowability of certain crude oil (e.g., waxy crude oil from Changqing Oilfield).

15 February 2024 in IEEE Transactions on Intelligent Vehicles
Verified
Journal Article
Closed access
ZRZhongyu Rao
Zhongyu Rao
YCYingfeng Cai
Yingfeng Cai
HWHai Wang
Hai Wang
YLYubo Lian
Yubo Lian
YZYilin Zhong
Yilin Zhong
YLYicheng Li
Yicheng Li
End to end autonomous driving system has rapidly progressed and garnered significant attention. Recently, multimodal fusion methods have surpassed vision-based methods in terms of performance. However, the multi-modal methods are both costly and time-confusion. This paper aims to propose a low-cost monocular end-to-end multi-task autonomous driving framework to address these challenges. In particular, to improve the performance of the vision-based model, our model considers three key aspects. First, to enhance the level of detail and visual representation of the surrounding environment, depth embeddings and semantic embeddings are utilized to assist in the perception of the Bird's Eye View(BEV) space. In addition, the motion queries are separated from the BEV queries and the novel transformer-based motion decoder is proposed to generate the output waypoints and target speed for motion planning. Subsequently, to optimize the utilization of the historic features, we introduced a streaming-based processing temporal module to fuse current and historic features. The performance of our proposed model is evaluated using the longest-06 and Town05-long benchmarks within the CARLA-0.9.10.1 simulator. The experiment results demonstrate that our proposed model performs state-of-the-art results on the longest-06 benchmark and promising results on the Town05-long benchmark even compared to these multi-modal sensor based models.

1 September 2025 in IEEE Transactions on Intelligent Transportation Systems
Verified
Journal Article
Closed access
YCYingfeng Cai
Yingfeng Cai
ZLZiheng Lu
Ziheng Lu
HWHai Wang
Hai Wang
YLYubo Lian
Yubo Lian
LCLong Chen
Long Chen
Qingchao LiuQingchao Liu
Qingchao Liu
The representation form of multimodal motion prediction results is critical to the efficiency of the prediction-planning workflow. However, most existing methods generate multiple sets of trajectories for each independent agent to cover diverse modalities, overlooking interactions among agents during the prediction period, making it challenging to achieve scene-level consistency in multi-agent predictions. This paper introduces MliG, a scene-level multimodal conditional motion prediction framework based on Multi-Layer Interaction Graph. The interaction graph designed explicitly represent the interaction behaviors of multi-agents over future periods, guiding conditional trajectory prediction to achieve planning-friendly prediction result representation. First, an Interaction-Mask-based determination strategy is introduced to obtain ground truth interaction labels, enabling data-driven implicit relationship learning, and the generated training labels can support the accurate construction of the interaction graph. Second, scene simplification and modality decoupling using the interaction graph make it easier to fuse information and represent driving scenarios efficiently. Based on the multi-layer design, the complex interactions between agent pairs are decomposed into different group and modality layers, ensuring the diversity of interactions while improving the efficiency of conditional predictions. Extensive experiments on the Argoverse 2, Argoverse 1, and INTERACTION Datasets demonstrate that the proposed MliG achieves high prediction accuracy and significantly improves the scene-level consistency of trajectory prediction modalities for multi-agents in the same driving scenario.

3 March 2025 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
ZLZiheng Lu
Ziheng Lu
YCYingfeng Cai
Yingfeng Cai
HWHai Wang
Hai Wang
YLYubo Lian
Yubo Lian
LCLong Chen
Long Chen
XSXiaoqiang Sun
Xiaoqiang Sun
Due to the inherent uncertainty within agents, multimodal motion prediction is required in real-world scenarios. Mainstream algorithms employ anchors or prior candidates to achieve multimodal predictions, and the primary challenge is to enhance the rationality and diversity of the modality distribution while ensuring prediction accuracy. This paper presents a progressive modality construction and refinement strategy incorporating the constraints of basic behaviors for multimodal motion prediction. Firstly, different generalized modality regions are partitioned based on high-level control commands, incorporating the constraints from basic driving behaviors. It can effectively aggregate training samples with similar driving behaviors, mitigating modal collapse and enhancing data utilization. Secondly, a hybrid-proposal refinement decoder is designed to integrate adaptive anchor-free queries with region-anchor corresponding to generalized modality region, further refining the diversity of agent behaviors. To fully leverage the advantages of the modality partition, region-aware attention and lane filtering strategies are proposed to balance prediction accuracy and efficiency effectively. Finally, validation on the Argoverse 1 and Argoverse 2 motion forecasting datasets demonstrates that the proposed model can achieve high prediction accuracy while efficiently utilizing training data to cover multimodal motion behaviors reasonably.