Lei Li

Lei Li

Auto-generated profile, based on publicly available information

Lei Li

School of Mechanical Engineering and Automation, Beihang University, Beijing, 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
26 August 2022 in Nature Communications
Verified
Journal Article
Open access
PDF file
WLWenbo Liu
Wenbo Liu
YDYouning Duo
Youning Duo
JLJiaqi Liu
Jiaqi Liu
FYFeiyang Yuan
Feiyang Yuan
LLLei Li
Lei Li
LWLi Wen
Li Wen
In this paper, we propose a multimodal flexible sensory interface for interactively teaching soft robots to perform skilled locomotion using bare human hands. First, we develop a flexible bimodal smart skin (FBSS) based on triboelectric nanogenerator and liquid metal sensing that can perform simultaneous tactile and touchless sensing and distinguish these two modes in real time. With the FBSS, soft robots can react on their own to tactile and touchless stimuli. We then propose a distance control method that enabled humans to teach soft robots movements via bare hand-eye coordination. The results showed that participants can effectively teach a self-reacting soft continuum manipulator complex motions in three-dimensional space through a “shifting sensors and teaching” method within just a few minutes. The soft manipulator can repeat the human-taught motions and replay them at different speeds. Finally, we demonstrate that humans can easily teach the soft manipulator to complete specific tasks such as completing a pen-and-paper maze, taking a throat swab, and crossing a barrier to grasp an object. We envision that this user-friendly, non-programmable teaching method based on flexible multimodal sensory interfaces could broadly expand the domains in which humans interact with and utilize soft robots.

25 May 2022 in Science Robotics
Verified
Journal Article
Closed access
LLLei Li
Lei Li
SWSiqi Wang
Siqi Wang
YZYiyuan Zhang
Yiyuan Zhang
SSShanyuan Song
Shanyuan Song
CWChuqian Wang
Chuqian Wang
LWLi Wen
Li Wen
Many real-world applications for robots—such as long-term aerial and underwater observation, cross-medium operations, and marine life surveys—require robots with the ability to move between the air-water boundary. Here, we describe an aerial-aquatic hitchhiking robot that is self-contained for flying, swimming, and attaching to surfaces in both air and water and that can seamlessly move between the two. We describe this robot’s redundant, hydrostatically enhanced hitchhiking device, inspired by the morphology of a remora ( Echeneis naucrates ) disc, which works in both air and water. As with the biological remora disc, this device has separate lamellar compartments for redundant sealing, which enables the robot to achieve adhesion and hitchhike with only partial disc attachment. The self-contained, rotor-based aerial-aquatic robot, which has passively morphing propellers that unfold in the air and fold underwater, can cross the air-water boundary in 0.35 second. The robot can perform rapid attachment and detachment on challenging surfaces both in air and under water, including curved, rough, incomplete, and biofouling surfaces, and achieve long-duration adhesion with minimal oscillation. We also show that the robot can attach to and hitchhike on moving surfaces. In field tests, we show that the robot can record video in both media and move objects across the air/water boundary in a mountain stream and the ocean. We envision that this study can pave the way for future robots with autonomous biological detection, monitoring, and tracking capabilities in a wide variety of aerial-aquatic environments.

11 March 2025 in IEEE Transactions on Pattern Analysis and Machine Intelligence
Verified
Journal Article
Closed access
LYLei Yang
Lei Yang
TTTao Tang
Tao Tang
JLJun Li
Jun Li
KYKun Yuan
Kun Yuan
KWKai Wu
Kai Wu
KYKaicheng Yu
Kaicheng Yu
While most recent autonomous driving system focuses on developing perception methods on ego-vehicle sensors, people tend to overlook an alternative approach to leverage intelligent roadside cameras to extend the perception ability beyond the visual range. We discover that the state-of-the-art vision-centric detection methods perform poorly on roadside cameras. This is because these methods mainly focus on recovering the depth regarding the camera center, where the depth difference between the car and the ground quickly shrinks while the distance increases. In this paper, we propose a simple yet effective approach, dubbed BEVHeight++, to address this issue. In essence, we regress the height to the ground to achieve a distance-agnostic formulation to ease the optimization process of camera-only perception methods. By incorporating both height and depth encoding techniques, we achieve a more accurate and robust projection from 2D to BEV spaces. On popular 3D detection benchmarks of roadside cameras, our method surpasses all previous vision-centric methods by a significant margin. In terms of the ego-vehicle scenario, BEVHeight++ surpasses depth-only methods with increases of +2.8% NDS and +1.7% mAP on the nuScenes test set, and even higher gains of +9.3% NDS and +8.8% mAP on the nuScenes-C benchmark with object-level distortion. Consistent and substantial performance improvements are achieved across the KITTI, KITTI-360, and Waymo datasets as well. The code is available at https://github.com/yanglei18/BEVHeight_Plus.}

20 September 2023 in Advanced Intelligent Systems
Verified
Journal Article
Open access
LLLei Li
Lei Li
WLWenbo Liu
Wenbo Liu
BTBocheng Tian
Bocheng Tian
PHPeiyu Hu
Peiyu Hu
WGWenzhuo Gao
Wenzhuo Gao
LWLi Wen
Li Wen
Hybrid aerial–aquatic robots can operate in both air and water and cross between these two. They can be applied to amphibious observation, maritime search and rescue, and cross‐domain environmental monitoring. Herein, an aerial–aquatic hitchhiking robot is proposed that can fly, swim, and rapidly cross the air–water boundaries (0.16 s) and autonomously attach to surfaces in both air and water. Inspired by the mechanoreceptors of the remora (Echeneis naucrates) disc, the robot's hitchhiking device is equipped with two flexible bioinspired tactile sensors (FBTS) based on a triboelectric nanogenerator for tactile sensing of attachment status. Based on tactile sensing, the robot can perform reattachment after leakage or adhesion failure, enabling it to achieve long‐term adhesion on complex surfaces. The rotor‐based aerial–aquatic robot, which has two thrust vectoring units for underwater locomotion, can maneuver to pitch, yaw, and roll 360° and control precision motion position. The field tests show that the robot can continuously cross the air–water boundary, attach to the rough stone surface, and record video in both air and underwater. This study may shed light on future autonomous robots capable of intelligent navigation, adhesion, and operation in complex aerial–aquatic environments.

2 January 2026 in Science Advances
Verified
Journal Article
Open access
BWBo Wang
Bo Wang
LLLei Li
Lei Li
MXMengfan Xu
Mengfan Xu
NHNannan Hu
Nannan Hu
WGWenzhuo Gao
Wenzhuo Gao
Junzhi YuJunzhi Yu
Junzhi Yu
Fish swim with four main gaits—anguilliform, subcarangiform, carangiform, and thunniform—produced by waves along varying portions of the body. However, how muscle activation length influences swimming performance remains poorly understood. We present a reconfigurable robotic fish that replicates all four gaits in a single platform by rapidly tuning its body stiffness. Vacuum-driven layer jamming muscles in four tensegrity joints enable quick (≤1 s) stiffness modulation (stiffness ratio of 46.6) and gait switching. In thunniform gait, the robot reaches 1.24 body lengths per second, whereas in anguilliform gait, it achieves agile maneuvering with a turning radius of 0.26 body lengths. Fluid simulations show that the thunniform gait generates stronger vortices and 142% more thrust compared with anguilliform motion at 5 Hz, explaining its high-speed performance. The robot dynamically adapts gaits during locomotion—using thunniform for fast traversal and anguilliform for obstacle negotiation—demonstrating environmental adaptability. This work advances understanding of aquatic multimodal locomotion.

13 July 2021 in Neural Processing Letters
Verified
Journal Article
Closed access
PDF file
LLLei Li
Lei Li
HZHaogang Zhu
Haogang Zhu
LWLongbo Wen
Longbo Wen
WLWeizhong Lan
Weizhong Lan
ZYZhikuan Yang
Zhikuan Yang