Although existing fire-fighting robots have been able to extinguish fires, the docking task of fire hoses has always been the key obstacle to achieving autonomous fire extinguishing functions for fire-fighting robots....
Although existing fire-fighting robots have been able to extinguish fires, the docking task of fire hoses has always been the key obstacle to achieving autonomous fire extinguishing functions for fire-fighting robots. therefore, it is crucial to enable robots to autonomously dock fire hoses to fire hydrants. To solve this problem, we design a new multi-DOF docking mechanism and develop a vision-guided autonomous fire hose docking system based on a robot operating system. When the robot detects the fire, the camera mounted on it can guide the docking system to complete the docking of the fire hose, laying the foundation for the robot to achieve autonomous fire extinguishing.
Underwater high-performance biological adhesion systems have attracted increasing research interest to inspire engineered adhesion systems. Switchable underwater adhesion systems typically employ suction adhesion. How...
详细信息
this paper presents a study on the optimization of You Only Look Once (YOLO) object detection algorithms for enhancing autonomous vehicle (AV) perception systems, with a focus on the United Arab Emirates (UAE) driving...
详细信息
ISBN:
(数字)9798331518301
ISBN:
(纸本)9798331518318
this paper presents a study on the optimization of You Only Look Once (YOLO) object detection algorithms for enhancing autonomous vehicle (AV) perception systems, with a focus on the United Arab Emirates (UAE) driving conditions. through evaluation of YOLO variants in simulated environments, YOLOv5m proved optimal by balanced detection accuracy and processing efficiency. the research involved creating a UAE-specific dataset to address regional scenarios, integrating real-time data processing in a virtual environment, and assessing algorithm performance across different conditions. Findings demonstrate the importance of varied environmental condition testing for object detection Algorithms, Furthermore the successful application of the custom UAE dataset underscores the necessity for regionally adapted training data in achieving high precision in object detection that is essential for navigation requirements of an AV.
Continuum robots are characterized by their high compliance and dexterity. However, these features also lead to a reduction in accuracy. While many literature on continuum robots address aspects such as design, kinema...
Continuum robots are characterized by their high compliance and dexterity. However, these features also lead to a reduction in accuracy. While many literature on continuum robots address aspects such as design, kinematics, and statics, there are few in-depth studies on the compliance of these robots. this paper introduces a numerical approach for obtaining the compliance of continuum robots, taking into account the influences of cable tensions, external loads, and gravity. Based on this method, an analysis to examine the compliance of a two-section cable-driven continuum robot within its workspace under various load conditions is conducted. Results indicate that the robot's compliance is correlated withthe external load and the shape of the robot. Moreover, the compliance of the robot in the x direction and y direction is highly related withthe distance between the robot's tip end and axes. these conclusions could be useful for the active compliance control of the robot.
In contrast to typical scenarios, orchards present a more intricate environment characterized by numerous irregular obstructions. this complexity often leads to the generation of zigzag paths when employing convention...
In contrast to typical scenarios, orchards present a more intricate environment characterized by numerous irregular obstructions. this complexity often leads to the generation of zigzag paths when employing conventional path planning algorithms. Moreover, the challenge is compounded by the orchard's slippery terrain and the fact that traditional path planning approaches do not consider robot kinematics, necessitating abrupt changes in acceleration. Such conditions make it arduous for orchard robots to follow these paths effectively. To mitigate these issues, this study introduces the minimum Jerk principle, which focuses on minimizing the rate of change in robot acceleration, as the cost function for optimizing the paths devised by standard path planning methods. the paper culminates in an empirical evaluation using a robot in an orchard setting. Experimental results demonstrate a substantial reduction in the robot's angular velocity fluctuations and a 12% decrease in tracking time, compared to paths generated by conventional methods.
In order to improve the path tracking accuracy and stability of low-speed autonomous vehicles during turning maneuvers, this study proposes an adaptive forward-looking distance path tracking algorithm based on road cu...
In order to improve the path tracking accuracy and stability of low-speed autonomous vehicles during turning maneuvers, this study proposes an adaptive forward-looking distance path tracking algorithm based on road curvature. the algorithm utilizes vehicle localization data and the curvature information of the path at the look_ahead point. By combining a formula for adaptive forward-looking distance derived from extensive physical experiments, the algorithm adjusts the forward-looking distance to obtain appropriate look_ahead points and the steering angle of the autonomous vehicle's front wheels. While maintaining algorithm stability, it effectively reduces the lateral error of the autonomous vehicle during turning maneuvers. Experimental results demonstrate that the proposed adaptive forward-looking distance path tracking algorithm outperforms traditional pure tracking algorithms in terms of accuracy and stability.
Withthe rapid development of automatic driving technology, semantic segmentation of road image has become a hot research object of semantic segmentation. In order to better solve the problems of long image prediction...
Withthe rapid development of automatic driving technology, semantic segmentation of road image has become a hot research object of semantic segmentation. In order to better solve the problems of long image prediction time and rough category edges in road image segmentation, this paper proposes a segmentation network model based on deep separation and attention mechanism based on the optimization design of the backbone network of DeepLabV3+model. By improving the adaptability of the BNeck structure composed of DSC backbone network, reducing the number of model parameters to improve the prediction speed of the model, through the combination of spatial attention mechanism and channel attention mechanism, the weight of the model is more distributed among the regions of interest and channels, and finally using the overall architecture of encoder decoder to design a more efficient segmentation model.
Aiming to address the problem of robots based on QR code navigation being unable to reach QR code nodes in time to correct dead reckoning errors, this paper proposes an improved A* algorithm. the improved method first...
Aiming to address the problem of robots based on QR code navigation being unable to reach QR code nodes in time to correct dead reckoning errors, this paper proposes an improved A* algorithm. the improved method firstly groups all QR code nodes into a dedicated node list, separated from ordinary nodes. It then applies a QR code reward function to weight the nodes in the dedicated node list. Finally, it modifies the weighting of the heuristic function to prioritize generating paths that pass through specific nodes in order to reach the target node. through simulation experiments, the proposed improved method increases the number of times the robot passes through the QR code nodes while ensuring path safety, effectively reducing dead reckoning errors, and improving the accuracy and efficiency of robot navigation.
In response to the current situation of coal mine gas inspection and the shortcomings of traditional manual gas inspection and intelligent gas inspection, this paper introduces an implementation method of an unmanned ...
In response to the current situation of coal mine gas inspection and the shortcomings of traditional manual gas inspection and intelligent gas inspection, this paper introduces an implementation method of an unmanned coal mine gas inspection system. On the basis of constructing the lora wireless network in the coal mine underground, this method realizes a network communication method of local wireless network and underground backbone ring network; Considering that gas inspection sites in coal mines involve many communication and power supply cables that are difficult to lay, wireless multi-parameter inspection equipment, electronic fence equipment, etc. have been adopted to achieve automatic inspection and alarm; Finally, in conjunction withthe upper computer system software and mobile APP application, the fully automatic control process of gas inspection was completed, ensuring the authenticity and reliability of gas inspection data, greatly saving human resources investment on the coal mine site.
In recent studies on LiDAR SLAM, the achievement of robust optimized LiDAR odometry is the primary objective. For the mapping part, some studies focus on improving the processing of point cloud, while others aim to th...
详细信息
暂无评论