BYBodi Yuan

Bodi Yuan

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Bodi Yuan has a total of 18 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Actor-Hybrid-Attention-Critic for Multi-Logistic Robots Path Planning

    by Chunjie Yang, Bodi Yuan, Pengzhao Zhai

    With the proliferation of express delivery services, the demand for intelligent logistic services is increasing. More and more robots are operating simultaneously, which poses a great challenge to their extraction of key information from a large number of static and dynamic environments. To address this, a Multi Actor Hybrid Attention Critic (MAHAC) algorithm for path planning of multi-logistic robots is proposed. The reward function is improved, based on collision avoidance and accurate arrival at the target location, a power shortage and sparse penalty are added to improve the efficiency of path planning. A hybrid attention mechanism module is designed, with an imbedded Multi Layer Perceptron (MLP), to calculate the similarity between key and query. Combining the hybrid attention mechanism with the Multi Actor Attention Critic (MAAC) algorithm improves the ability of multi-logistic robots to extract critical states and action information in a variety of environments. In a simulated warehouse layout based on Amazon's Kiva system, the MAHAC algorithm is compared with MAAC, Multi-Agent Deep Deterministic Policy Gradient (MADDPG), and Multi-Actor Soft Actor-Critic (MASAC) algorithms. The MAHAC has a higher reward value, so as to select the better path in the least time, with the highest accuracy and no collisions, and hence is suitable for the path planning scenario of multi-logistic robots.


  • Open Access

    Allowing Safe Contact in Robotic Goal-Reaching: Planning and Tracking in Operational and Null Spaces

    by Xinghao Zhu, Wenzhao Lian, Bodi Yuan, C. Daniel Freeman, Masayoshi Tomizuka

    In recent years, impressive results have been achieved in robotic manipulation. While many efforts focus on generating collision-free reference signals, few allow safe contact between the robot bodies and the environment. However, in human's daily manipulation, contact between arms and obstacles is prevalent and even necessary. This paper investigates the benefit of allowing safe contact during robotic manipulation and advocates generating and tracking compliance reference signals in both operational and null spaces. In addition, to optimize the collision-allowed trajectories, we present a hybrid solver that integrates sampling- and gradient-based approaches. We evaluate the proposed method on a goal-reaching task in five simulated and real-world environments with different collisional conditions. We show that allowing safe contact improves goal-reaching efficiency and provides feasible solutions in highly collisional scenarios where collision-free constraints cannot be enforced. Moreover, we demonstrate that planning in null space, in addition to operational space, improves trajectory safety.

  • Open Access

    Semantically Robust Unpaired Image Translation for Data with Unmatched Semantics Statistics

    by Zhiwei Jia, Bodi Yuan, Kangkang Wang, Hong Wu, David Clifford, Zhiqiang Yuan, Hao Su

    Many applications of unpaired image-to-image translation require the input contents to be preserved semantically during translations. Unaware of the inherently unmatched semantics distributions between source and target domains, existing distribution matching methods (i.e., GAN-based) can give undesired solutions. In particular, although producing visually reasonable outputs, the learned models usually flip the semantics of the inputs. To tackle this without using extra supervision, we propose to enforce the translated outputs to be semantically invariant w.r.t. small perceptual variations of the inputs, a property we call "semantic robustness". By optimizing a robustness loss w.r.t. multi-scale feature space perturbations of the inputs, our method effectively reduces semantics flipping and produces translations that outperform existing methods both quantitatively and qualitatively.

  • Open Access

    Semantically Robust Unpaired Image Translation for Data with Unmatched Semantics Statistics

    by Zhiwei Jia, Bodi Yuan, Kangkang Wang, Hong Wu, David Clifford, Zhiqiang Yuan, Hao Su

    Many applications of unpaired image-to-image translation require the input contents to be preserved semantically during translations. Unaware of the inherently unmatched semantics distributions between source and target domains, existing distribution matching methods (i.e., GAN-based) can give undesired solutions. In particular, although producing visually reasonable outputs, the learned models usually flip the semantics of the inputs. To tackle this without using extra supervisions, we propose to enforce the translated outputs to be semantically invariant w.r.t. small perceptual variations of the inputs, a property we call "semantic robustness". By optimizing a robustness loss w.r.t. multi-scale feature space perturbations of the inputs, our method effectively reduces semantics flipping and produces translations that outperform existing methods both quantitatively and qualitatively.