The Research And Design Of Home Service Robot For The Olders

As the aging of population is deepening in China,demand for home service robot for the olders is growing.And the government pay attention to study for home service robot for the olders and increasing the input to it constantly.So,home service robot for the olders has received attention by many researchers and scholars.Correlation technique of home service robot for the olders become hot topic.For example,man-machine interaction,path planning,item identification and SLAM.In this paper,offline speech recognition and global path planning are regarded as the key research object.In the limited scenes of life assisted services for the olders,the aged can control robot to complete service by voice commands.The related algorithms of speech recognition and language model training tool are thoroughly studied in this paper.In order to overcome low recognition rate while using raw language model,small vocabulary corpus is built and language model is trained according to requirements.And keyword recognition is applied to decoding process of some phrases,which reduces the false accept rate to some ways.Then based on the study,a offline speech recognition system based on Pocketsphinx is designed and completed.This system is installed in Android smartphone to test performances.The experimental results show that correct recognition accuracy is about 97%by three different testers.Aiming at the problem of many redundant path points,large cumulative turning angle and the robot not being able to its posture at the path point in Jump Point Research,an improved algorithm is designed.In this improved algorithm,the connectivity between the previous point and the last point is checked,and vector cross product and the vector dot product are applied to identify the direction and calculate the rotation angle.The experimental results show that in the map size of 30×30,the total path length decreases by about 30%and the cumulative turning angle decreases by about 50%in improved algorithm,which improves efficiency of path planning.The improved jump point search algorithm proposed in this paper generates final path with shorter path length,fewer path points.At the same time,the robot can autonomously adjust the pose at the point.Through the study and design,home service robot for olders prototype system is completed,and the improved algorithm is applied to path planning.In this system,robot is controlled by voice command and the basic function of indoor navigation is achieved.