Hongru Tang

Dynamic hand gesture recognition based on SURF tracking

A novel method of dynamic hand gesture recognition based on Speeded Up Robust Features (SURF) tracking is proposed. The main characteristic is that the dominant movement direction of matched SURF points in adjacent frames is used to help describing a hand trajectory without detecting and segmenting the hand region. The dynamic hand gesture is then modeled by a series of trajectory direction data streams after time warping. Accordingly, the data stream clustering method based on correlation analysis is developed to recognize a dynamic hand gesture and to speed up calculation. The proposed algorithm is tested on 26 alphabetical hand gestures and yields a satisfactory recognition rate which is 87.1% on the training set and 84.6% on the testing set.

Dynamic Hand Gesture Recognition Based on SURF Tracking: Dynamic Hand Gesture Recognition Based on SURF Tracking

A novel method of dynamic hand gesture recognition based on Speeded Up Robust Features (SURF) tracking is proposed. The main characteristic is that the dominant movement direction of matched SURF points in adjacent frames is used to help describing a hand trajectory without detecting and segmenting the hand region. The dynamic hand gesture is then modeled by a series of trajectory direction data streams after time warping. Accordingly, the data stream clustering method based on correlation analysis is developed to recognize a dynamic hand gesture and to speed up calculation. The proposed algorithm is tested on 26 alphabetical hand gestures and yields a satisfactory recognition rate which is 87.1% on the training set and 84.6% on the testing set.

Research on Centroid Position for Stairs Climbing Stability of Search and Rescue Robot

This paper represents the relationship between the stability of stairs climbing and the centroid position of the search and rescue robot. The robot system is considered as a mass point- plane model and the kinematics features are analyzed to find the relationship between centroid position and the maximal pitch angle of stairs the robot could climb up. A computable function about this relationship is given in this paper. During the stairs climbing, there is a maximal stability-keeping angle depends on the centroid position and the pitch angle of stairs, and the numerical formula is developed about the relationship between the maximal stability-keeping angle and the centroid position and pitch angle of stairs. The experiment demonstrates the trustworthy and correction of the method in the paper.

A Combination of Terrain Prediction and Correction for Search and Rescue Robot Autonomous Navigation

This paper presents a novel two-step autonomous navigation method for search and rescue robot. The algorithm based on the vision is proposed for terrain identification to give a prediction of the safest path with the support vector regression machine (SVRM) trained off-line with the texture feature and color features. And correction algorithm of the prediction based the vibration information is developed during the robot traveling, using the judgment function given in the paper. The region with fault prediction will be corrected with the real traversability value and be used to update the SVRM. The experiment demonstrates that this method could help the robot to find the optimal path and be protected from the trap brought from the error between prediction and the real environment.

Human-robot collaborative teleoperation system for semi-autonomous reconnaissance robot

Several key issues about the teleoperation system for semi-autonomous reconnaissance robot have been addressed. A human-robot collaborative semi-autonomous mobile robot architecture (SAMRA) which combine the complementary capabilities of both robots local autonomy and human intelligence is proposed firstly. Then, a novel autonomous mission executive mechanism based on macro behavior and a hybrid behavior coordination mechanism is brought out. Meanwhile, several types of intelligent behavior are defined briefly. The implement method of teleoperation system software based on multi-agent system is introduced in succession. Finally, a visualized human-robot interface with live video/audio, 3D drawing simulation, and graphic mission planner is presented. Experimental results demonstrate the autonomous mission executor is able to interpret and execute the planned mission specification successfully, and the human-robot collaborative teleoperation system for semi-autonomous reconnaissance robot has excellent property of telepresence, flexibility and robustness.

Combining Vision Learning and Interaction for Mobile Robot Path Planning

This paper addresses the question of how to make a robot learn natural terrain selectively and use the knowledge to estimate the terrain for planning an optimal path. A scheme which combines vision learning and interaction is proposed. The vision learning module employs an online boosting learning algorithm to constantly receive and learn the terrain samples each of which comprise the visual features extracted from the sub terrain region image and the traversability measured by the onboard Inertia Measurement Unit (IMU). Using this knowledge, the robot could estimate the new terrains and search for the optimal path to travel using the particle swarm optimization method. To overcome the shortcoming that the robot could not understand the intricate environment exactly, the vision interaction method, which complements the robots capacity of terrain estimation with the human reasoning ability of path correction, is further applied. Experimental results show the effectiveness of the proposed method.

Research on Navigation for Search and Rescue Robot Based on Traversability

This paper presents a navigation algorithm based on traversability for search and rescue robot. The traversability includes pre-traversability and real-time traversability. The pre-traversability is calculated using terrain roughness and roll and pitch change. The terrain roughness is based on the texture information from the image captured by the on-board camera. An artificial neural network has been trained to find the relationship between roll and pitch changes and image characters. In order to avoid the localization that the intransitable region could not be completely identified just using the image process method, the real-time traversability with slip element has been adopted to correct the error in pre-traversability. The fuzzy logic approach has been developed for the traversability integration. The experiment demonstrates that the navigation algorithm is effective.