To improve the performance of the image semantic segmentation algorithm and make the algorithm achieve a better balance between accuracy and real-time performance when segmenting images, this paper proposes a real-time image semantic segmentation model based on an improved DeepLabv3+ network. First, the MobileNetV2 model with less computational overhead and number of parameters is selected as the backbone network to improve the segmentation speed; then, the Feature Enhancement Module (FEM) is introduced to several shallow features with different scale sizes in MobileNetV2, and then these shallow features are fused to improve the utilization rate of the model encoder on the edge information, to retain more detailed information and to improve the network’s feature representation ability for complex scenes; finally, to address the problem that the output feature maps of Atrous Spatial Pyramid Pooling (ASPP) module do not pay enough attention to detailed information after merging, the FEM attention mechanism is introduced on the feature maps processed by the ASPP module. The algorithm in this study achieves 76.45% for mean intersection over union (mIoU) accuracy with 29.18 FPS real-time performance in the PASCAL VOC2012 Augmented dataset; and 37.31% mIoU accuracy with 23.31 FPS real-time performance in the ADE20K dataset. The experimental results show that the algorithm in this study achieves a good balance between accuracy and real-time performance, and its image semantic segmentation performance is significantly improved compared to DeepLabv3+ and other existing algorithms.

In view of the large randomness, redundant path nodes, and low search efficiency of RRT-connect in a complex obstacle environment, this study intends to develop a path-planning method combining RRT-connect and Informed RRT*. First, to solve the problem of large sampling randomness, the Informed RRT* algorithm is combined to adopt a simpler rectangle and limit the sampling range to the rectangle. Second, for the poor quality of the search path, the dynamic step size is used for growth extension, the reverse greedy algorithm is used to delete redundant nodes, the spline curve is used to smooth the path such that the position meets the cubic spline curve and the speed meets the quadratic spline curve, and the final path is optimized. Finally, the proposed algorithm is verified in the simulation and real world using a self-developed quadrupedal robot. Compared with the original RRT-connect algorithm, the first solution time, total number of nodes, and initial path cost were reduced by more than 11%, 8.5%, and 2.5%, respectively.