Our approach begins with a voxel-based semantic segmentation (SS) pretraining task, designed to explore the semantics and geometry of incomplete regions while acquiring transferable meta-knowledge. Using simulated cooperative perception datasets, we supervise the training of a single vehicle’s perception using the aggregated sensor data from multiple nearby connected autonomous vehicles (CAVs), generating richer and more comprehensive labels. This meta-knowledge is then adapted to the target domain through a dual-phase training strategy—without adding extra model parameters—ensuring efficient deployment. To further enhance the model’s ability to capture long-sequence relationships in 3D voxel grids, we integrate Mamba blocks with deformable convolution and large-kernel attention into the backbone network. Extensive experiments show that MetaSSC achieves state-of-the-art performance, surpassing competing models by a significant margin, while also reducing deployment costs. 

‘NodeSet_2D.mat’ and ‘LinkSet_2D.mat’ describe the node set and link set in Wangjing road network, and ‘ODset.mat’ describe the O-D pairs of demand for UAVs.

This study utilizes datasets from the original baselines, including Los-loop & SZ-taxi (https://github.com/lehaifeng/T-GCN), PEMS-BAY and METR-LA (https://github.com/liyaguang/DCRNN), and PeMSD4 and PeMSD8 (https://github.com/wanhuaiyu/ASTGCN#datasets). The code is available at https://github.com/Chengfeng-Jia/SAFER-Predictor. 

All the materials source coding and data of the Paper titled a multi-agent social interaction model for autonomous vehicle testing

Replication Package for "Towards robust motion control in multi-source uncertain scenarios by robust policy iteration"