A siamese network-based approach for vehicle pose estimation

Zhao, Heng; Tao, Bo; Huang, Licheng; Chen, Baojia

doi:10.3389/fbioe.2022.948726

Cited by 4 publications

(3 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Considering that the orthogonality constraint is difficult to realize in the deep learning process, the Euler angles, which are more suitable as a representation of the regression results of deep learning, are used instead of the rotation matrix. Euler angles α, β, γ around X, Y, Z axes its process is shown in equation (12).…”

Section: Implementation Detailsmentioning

confidence: 99%

See 1 more Smart Citation

Self-supervised monocular visual odometry based on cross-correlation

Hu,

Tao,

Qian

et al. 2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

Visual odometry constitutes a critical component in enabling autonomous navigation. However, the existing methods are limited by the feature extraction and matching accuracy, and cannot show good real-time performance while combining accuracy and robustness. In this paper, we propose a novel monocular visual odometry framework based on cross-correlation. The framework starts with a parameter-sharing Siamese network to build feature extractors that can simultaneously process multiple images as inputs. Moreover, we design cross-correlation modules and define a cross-correlation matrix to describe the strength of correlation between different parts of the input feature maps, reflecting the rotational and translational transformations of the input images. Furthermore, a novel loss function is introduced to impose constraints on the network. Additionally, a fully convolutional network is designed for pose estimation, computing poses alterations from the structure of the cross-correlation matrix. Channel attention and spatial attention mechanisms are introduced to improve the performance. More importantly, our method innovatively uses time intervals as labels, enables self-supervised training, and relies only on a monocular camera. Experimental results on the KITTI visual odometry dataset show that our method produces competitive performance, demonstrating the superiority of the proposed method.

show abstract

Section: Implementation Detailsmentioning

confidence: 99%

“…Generally, deep learning methods can be categorized as supervised learning, unsupervised learning, and self-supervised learning. Among them, the exploration of supervised learning methods [11,12] has been more in-depth. However, annotating the escalating volume of data across diverse scenarios entails substantial expenses.…”

Section: Introductionmentioning

confidence: 99%

Self-supervised monocular visual odometry based on cross-correlation

Hu,

Tao,

Qian

et al. 2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…Some scholars have improved the control accuracy by combining the neural network method with the dynamic model. [72][73][74][75][76][77][78] Meng et al 79 investigated an uncertain dual-linked rigid-flexible robotic arm with amplitude constraints, aiming at its position control through motion planning and adaptive tracking with radial basis function NN and SMC. Dai et al 80 proposed a robust model predictive control algorithm for time-varying trajectory tracking of manipulator.…”

Section: Related Workmentioning

confidence: 99%

Manipulator trajectory tracking based on adaptive fuzzy sliding mode control

Zhao

Tao

et al. 2023

Concurrency and Computation

Self Cite

View full text Add to dashboard Cite

Sliding mode control is one of the common control methods for manipulators, but the discontinuity of sliding mode control can cause jitter and vibration of the manipulator system. This article takes the Dobot Magician manipulator as the research object and constructs a simplified model of the manipulator dynamics. And the adaptive fuzzy sliding mode control method is designed in combination with fuzzy control theory, which effectively solves the jitter problem of the control torque. In the MATLAB/Simulink simulation environment analysis show that the proposed adaptive fuzzy sliding mode control method solves the jitter problem in sliding mode control with good stability, robustness and tracking performance. By combining the adaptive fuzzy sliding mode control method with the manipulator and comparing the motion time of the same motion trajectory with the PID control method in Dobot Studio, the hardware experiment results effectively verify the feasibility and effectiveness of the designed control method.

show abstract

Research on Vehicle Pose Detection Method Based on a Roadside Unit

Ni,

Kong,

Yan

et al. 2024

Sensors

View full text Add to dashboard Cite

Vehicle pose detection plays a vital role in modern automotive technology, which can improve driving safety, enhance vehicle stability and provide important support for the development of autonomous driving technology. The current pose estimation methods have the problems of accumulation errors, large algorithm computing power, and expensive cost, so they cannot be widely used in intelligent connected vehicles. This paper proposes a vehicle pose detection method based on an RSU (Roadside Unit). First, the on-board GPS performs the positioning of the target vehicle and transmits the positioning information to the RSU through the UDP (User Data Protocol). Next, the RSU transmits a forward command to the OBU (On-board Unit) through the UDP. The OBU sends the command to the ECU (Electronic Control Unit) to control the vehicle forward. Then, the RSU detects and tracks the vehicle. The RSU takes pictures of two images before and after the movement and obtains the coordinates of the four angle points and the center point by image processing. The vehicle heading direction is determined by the moving direction of the center point of the front and rear two images. Finally, the RSU captures the vehicle images in real time, performs the process of tracking, rectangular fitting and pose calculation to obtain the pose information and transmits the information to the OBU to complete the whole process of vehicle pose detection and information transmission. Experiments show that the method can realize accurate and efficient detection of vehicle pose, meet the real-time requirements of vehicle pose detection, and can be widely used in intelligent vehicles.

show abstract

A siamese network-based approach for vehicle pose estimation

Cited by 4 publications

References 46 publications

Self-supervised monocular visual odometry based on cross-correlation

Self-supervised monocular visual odometry based on cross-correlation

Manipulator trajectory tracking based on adaptive fuzzy sliding mode control

Research on Vehicle Pose Detection Method Based on a Roadside Unit

Contact Info

Product

Resources

About