Baoqian Wang (王宝乾)

I am an Intelligent Systems Engineer at The Boeing Company, working on reinforcement learning for robot decision making problems.

In 2023, I obtained my Ph.D. degree in Electrical and Computer Engineering jointly from University of California San Diego and San Diego State University, advised by Prof. Junfei Xie and Prof. Nikolay Atanasov. I received my M.S. degree in Computer Science from Texas A&M University Corpus Christi in 2019 and B.S. degree in Prospecting Technology and Engineering from Yangtze University, China in 2017.

Email  /  Google Scholar  /  Github  /  LinkedIn

DARL1N: Distributed multi-Agent Reinforcement Learning with One-hop Neighbors
Baoqian Wang, Junfei Xie, Nikolay A.Atanasov
2022 IEEE/RSJ International Conference on Intelligent Robots and Systems
Abstract: In this paper, introduce a scalable MARL method called Distributed multi-Agent Reinforcement Learning with One-hop Neighbors (DARL1N). DARL1N is an off-policy actor-critic method that breaks the curse of dimensionality by decoupling the global interactions among agents and restricting information exchanges to one-hop neighbors. Each agent optimizes its action value and policy functions over a one-hop neighborhood, significantly reducing the learning complexity, yet maintaining expressiveness by training with varying numbers and states of neighbors. This structure allows us to formulate a distributed learning framework to further speed up the training procedure.

On Batch-Processing Based Coded Computing for Heterogeneous Distributed Computing Systems
Baoqian Wang, Junfei Xie, Kejie Lu, Yan Wan, Shengli Fu
IEEE Transcations on Network Science and Engineering, 2021
Abstract: Coded distributed computing (CDC) can efficiently facilitate many delay-sensitive computation tasks against unexpected latencies in distributed computing systems. In this paper, we focus on practical computing systems with heterogeneous computing resources, and design a novel CDC approach, called batch-processing based coded computing (BPCC), which exploits the fact that every computing node can obtain some coded results before it completes the whole task. To this end, we first describe the main idea of the BPCC framework, and then formulate an optimization problem for BPCC to minimize the task completion time by configuring the computation load.

Coding for Distributed Multi-Agent Reinforcement Learning
Baoqian Wang, Junfei Xie, Nikolay A.Atanasov,
2021 International Conference on Robotics and Automation (ICRA)
Abstract: This paper aims to mitigate straggler effects in synchronous distributed learning for multi-agent reinforcement learning (MARL) problems. Stragglers arise frequently in a distributed learning system, due to the existence of various system disturbances such as slow-downs or failures of compute nodes and communication bottlenecks. To resolve this issue, we propose a coded distributed learning framework, which speeds up the training of MARL algorithms in the presence of stragglers, while maintaining the same accuracy as the centralized approach. As an illustration, a coded distributed version of the multi-agent deep deterministic policy gradient(MADDPG) algorithm is developed and evaluated. Different coding schemes, including maximum distance separable (MDS)code, random sparse code, replication-based code, and regular low density parity check (LDPC) code are also investigated. Simulations in several multi-robot problems demonstrate the promising performance of the proposed framework.

Data-Driven Multi-UAV Navigation in Large-Scale Dynamic Environments Under Wind Disturbances
Baoqian Wang, Junfei Xie, Jun Chen
AIAA Scitech 2021 Forum
Abstract: In the near future, large amount of unmanned aerial vehicles (UAVs) are expected to appear in the airspace. To ensure the safety of the airspace, there are many daunting technical problems to tackle, one of which is how to navigate multiple UAVs safely and efficiently in the large-scale airspace with both static and dynamic obstacles under wind disturbances. This paper solves this problem by developing a novel data-driven multi-UAV navigation framework that combines $A^*$ algorithm with a state-of-the-art deep reinforcement learning (DRL) method. The $A^*$ algorithm generates a sequence of waypoints for each UAV and the DRL ensures that the UAV can reach each waypoint in order while satisfying all dynamic constraints and safety requirements. Furthermore, our framework significantly expedites the online planning procedure by offloading most computations to offline and limiting online computing to only path fine-tuning and dynamic obstacle avoidance. The simulation studies show the good performance of the proposed framework.

Computing in the air: An Open Airborne Computing Platform
Baoqian Wang, Junfei Xie, Songwei Li, Yan Wan, Yixin Gu, Shengli Fu, Kejie Lu,
IET Communications, 2020
Abstract: In this study, we aim to design an open UAS-based airborne computing platform with advanced onboard computing capability. This platform was designed from three aspects: hardware, software, and applications. In particular, feasible computing hardware to perform UAS onboard computing is first considered and a prototype is then designed. To enhance the flexibility and programmability of the platform, two key virtualisation techniques are then investigated. Finally, they test the performance of their prototype by executing real UAS onboard computing tasks, the results of which verify the feasibility and potentials of the proposed airborne computing platform.

3-D Trajectory Modeling for Unmanned Aerial Vehicles
Baoqian Wang, Junfei Xie, Yan Wan, GA Guijarro Reyes, LR Garcia Carrillo,
AIAA Scitech 2019 Forum
Abstract: This paper aims to develop a hybrid 3-dimensional (3-D) UAV trajectory modeling framework, which integrates the physically-based and data-based models (LSTM-based models).The key idea is to use a physically-based model, which may not perfectly capture the true dynamics of the UAV of interest, to generate large amount of trajectory data and use these data to train a data-based model.This baseline model is then tuned using small amount of real flight data to capture the true dynamics of the targeted UAV.

Vision-based Autonomous Driving Robot Capable of Navigation in Unknown and Dynamic Rural Environments
Ramiz Hanan, David Pierce Walker-Howell, Leo Peralta, Junfei Xie, Baoqian Wang,
2020 IEEE/ASME International Conference on Advanced Intelligent Mechatronics
Abstract: In this project, we developed a modular, intelligent, and autonomous driving robot that is not only capable of navigating in a known urban environment, but also in an unknown and dynamic rural environment with unpaved roads, which is achieved by a deep reinforcement learning algorithm, namely GA3C.

Geolocation using video sensor measurements
Abstract: In this project, a simple geolocation algorithm using video sensor measurements is implemented in real robot. In particular, the mobile robot positions in the image frame are obtained through YoloV3. Given the position of the camera in real world and intrinsic parameters of the camera, the position of the mobile robot in the real world frame can be obtained through camera projection model. The estimated position are then compared to the real positions recorded by motion capture system.

Pocket Planner
Abstract: In project, we developed an Andriod App called Pocket Planner, which aims to manage the time efficiently by allowing users to add their daily events and schedules in a flexiable and easy way. The code is written in Android Java.

This website is built by using template from Jon Barron