AZAlex Zhou

Alex Zhou

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

Alex Zhou has a total of 49 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    SEER: Safe Efficient Exploration for Aerial Robots using Learning to Predict Information Gain

    by Yuezhan Tao, Yuwei Wu, Beiming Li, Fernando Cladera, Alex Zhou, Dinesh Thakur, Vijay Kumar

    We address the problem of efficient 3-D exploration in indoor environments for micro aerial vehicles with limited sensing capabilities and payload/power constraints. We develop an indoor exploration framework that uses learning to predict the occupancy of unseen areas, extracts semantic features, samples viewpoints to predict information gains for different exploration goals, and plans informative trajectories to enable safe and smart exploration. Extensive experimentation in simulated and real-world environments shows the proposed approach outperforms the state-of-the-art exploration framework by 24% in terms of the total path length in a structured indoor environment and with a higher success rate during exploration.

  • Open Access

    Abstract 324: Discovery of novel functional TROP2 antibodies for treatment of epithelial cancers

    by Israel Matos, Yahya Ashraf, Hiba Zahreddine, Liying Gong, Amanda Baumholtz, Alex Zhou, Claire Bonfils, Aniel Moya-Torres, Yun Cui, Xiaowei Wang, et al.

    Trophoblast cell surface antigen-2 (TROP2) is a membrane-bound protein expressed during development and at lower levels in normal adult tissues. TROP2 has pleiotropic functions including formation of cell-cell junctions by mediating interactions with extracellular and intracellular proteins. In situations of TROP2 deficiency its role in normal tissues appears to be compensated by expression of EpCAM, a highly related family member. In cancer, the levels of TROP2 are increased in many epithelial tumor types. Overexpression of TROP2 confer oncogenic behavior in several in vitro and in vivo models. Both its extracellular and intracellular domains play a role in modulating interactions with ligands and/or receptors and with intracellular signaling partners to result in an oncogenic phenotype. In many solid tumors TROP2 upregulation is associated with increased tumor aggressiveness, metastasis, and decreased survival in large groups of difficult-to-treat cancers, making it an attractive target for cancer therapy. TROP2-based ADCs have been developed and one has been approved for therapeutic use in triple negative breast cancer and advanced bladder cancer. The toxicity associated with the toxin payload limits their use including in combination with chemotherapy in the earlier lines of treatment, and by the development of resistance to the payload. We isolated several anti-TROP2 heavy chain-only antibodies (HCAb) by immunization of a proprietary transgenic HCAb mouse. In this abstract, we describe the properties and activity of several of these antibodies. These antibodies are characterized by potent in vivo antitumor activity in established tumor models in NCG mice in the absence of ADCC, and by the inhibition of tumor cell viability in vitro, directly and in the absence of any effector cells. In NCG mice, which are characterized by a short antibody half-life, anti-TROP2 HCAbs displayed a T1/2 of approximately 3.5 days. These antibodies displayed very potent anti-tumor activity in established tumor xenograft models, causing tumor regression at 2 mg/kg and 80% TGI at 0.5 mg/kg in NCG mice when administered once per week. They also caused tumor regression when dosed once every three weeks in NCG mice and in tumors of very large sizes. Therefore, these antibodies directly target one or more aspects of TROP2 tumor biology that have a profound and long-lasting effect on the cancer cells. Upon glycoengineering, the afucosylated version of the antibodies displayed a strongly enhanced ADCC activity. In vitro they are characterized by different internalization rates which define several non-overlapping epitopes on TROP2. They are characterized by picomolar binding affinity to recombinant human TROP2 as determined by SPR. These antibodies bind with greater affinity to human and cyno but bind poorly to mouse, rat and rabbit. These potent naked TROP2 antibodies represent new opportunities to address TROP2 expressing cancers. Citation Format: Israel Matos, Yahya Ashraf, Hiba Zahreddine, Liying Gong, Amanda Baumholtz, Alex Zhou, Claire Bonfils, Aniel Moya-Torres, Yun Cui, Xiaowei Wang, Richard Wargachuk, Tiffany Cheng, Elijus Undzys, Shugang Yao, Jacynthe Linda Tolouse, Dominic Hou, Gordon Ngan, Luis daCruz, David Young. Discovery of novel functional TROP2 antibodies for treatment of epithelial cancers [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 324.

  • Top Cited Papers

    Large-Scale Autonomous Flight With Real-Time Semantic SLAM Under Dense Forest Canopy

    by Xu Liu, Guilherme V. Nardari, Fernando Cladera, Yuezhan Tao, Alex Zhou, Thomas Donnelly, Chao Qu, Steven W. Chen, Roseli A. F. Romero, Camillo J. Taylor, et al.

    Semantic maps represent the environment using a set of semantically meaningful objects. This representation is storage-efficient, less ambiguous, and more informative, thus facilitating large-scale autonomy and the acquisition of actionable information in highly unstructured, GPS-denied environments. In this letter, we propose an integrated system that can perform large-scale autonomous flights and real-time semantic mapping in challenging under-canopy environments. We detect and model tree trunks and ground planes from LiDAR data, which are associated across scans and used to constrain robot poses as well as tree trunk models. The autonomous navigation module utilizes a multi-level planning and mapping framework and computes dynamically feasible trajectories that lead the UAV to build a semantic map of the user-defined region of interest in a computationally and storage efficient manner. A drift-compensation mechanism is designed to minimize the odometry drift using semantic SLAM outputs in real time, while maintaining planner optimality and controller stability. This leads the UAV to execute its mission accurately and safely at scale.


  • Mine Tunnel Exploration Using Multiple Quadrupedal Robots

    by Ian D. Miller, Fernando Cladera, Anthony Cowley, Shreyas S. Shivakumar, Elijah S. Lee, Laura Jarin-Lipschitz, Akhilesh Bhat, Neil Rodrigues, Alex Zhou, Avraham Cohen, et al.

    Robotic exploration of underground environments is a particularly challenging problem due to communication, endurance, and traversability constraints which necessitate high degrees of autonomy and agility. These challenges are further exacerbated by the need to minimize human intervention for practical applications. While legged robots have the ability to traverse extremely challenging terrain, they also engender new challenges for planning, estimation, and control. In this work, we describe a fully autonomous system for multi-robot mine exploration and mapping using legged quadrupeds, as well as a distributed database mesh networking system for reporting data. In addition, we show results from the DARPA Subterranean Challenge (SubT) Tunnel Circuit demonstrating localization of artifacts after traversals of hundreds of meters. These experiments describe fully autonomous exploration of an unknown Global Navigation Satellite System (GNSS)-denied environment undertaken by legged robots.


  • Open Access

    MAVNet: an Effective Semantic Segmentation Micro-Network for MAV-based Tasks

    by Ty Nguyen, Shreyas S. Shivakumar, Ian D. Miller, James Keller, Elijah S. Lee, Alex Zhou, Tolga Ozaslan, Giuseppe Loianno, Joseph H. Harwood, Jennifer Wozencraft, et al.

    Real-time semantic image segmentation on platforms subject to size, weight and power (SWaP) constraints is a key area of interest for air surveillance and inspection. In this work, we propose MAVNet: a small, light-weight, deep neural network for real-time semantic segmentation on micro Aerial Vehicles (MAVs). MAVNet, inspired by ERFNet, features 400 times fewer parameters and achieves comparable performance with some reference models in empirical experiments. Our model achieves a trade-off between speed and accuracy, achieving up to 48 FPS on an NVIDIA 1080Ti and 9 FPS on the NVIDIA Jetson Xavier when processing high resolution imagery. Additionally, we provide two novel datasets that represent challenges in semantic segmentation for real-time MAV tracking and infrastructure inspection tasks and verify MAVNet on these datasets. Our algorithm and datasets are made publicly available.