| Seznam |
| Téma: | Navigace robotu s přihlédnutím k charakteru scény |
|---|---|
| Katedra: | Katedra kybernetiky |
| Vedoucí: | prof. Ing. Tomáš Svoboda, Ph.D. |
| Vypsáno jako: | Diplomová práce, Bakalářská práce, Semestrální projekt |
| Popis: | In many robotic applications, the robot goal is to follow a prescribed path close enough. An outdoor deployment can use a sequence of GPS-defined waypoints. A standard geometric cost function penalises distance from the ideal path. However, the environment may be too complex to follow the path exactly. Driving on a public road, street or sidewalk could serve as an example. Semantic scene segmentation shall provide another cost function to be combined with the geometric one.
Initial phase of the project should focus on driving on public roads - stable driving on a roadside. The scene segmentation may include multimodal data. |
| Literatura: | Jingdong Wang, Ke Sun, Tianheng Cheng, Borui Jiang, Chaorui Deng, Yang Zhao, Dong Liu, Yadong Mu, Mingkui Tan, Xinggang Wang, Wenyu Liu, and Bin Xiao. Deep High-Resolution Representation Learning
for Visual Recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), March 2020. https://github.com/HRNet/HRNet-Semantic-Segmentation Jiacong Xu and Zixiang Xiong and Shankar P. Bhattacharyya. A Real-time Semantic Segmentation Network Inspired from PID Controller. arXiv.2206.02066, https://arxiv.org/abs/2206.02066 https://github.com/XuJiacong/PIDNet Awet Haileslassie Gebrehiwot, Patrik Vacek, David Hurych, Karel Zimmermann, Patrick Perez, Tomáš Svoboda. Teachers in concordance for pseudo-labeling of 3D sequential data. https://doi.org/10.48550/arXiv.2207.06079, https://github.com/ctu-vras/t-concord3d |