SparseDriveV2 end-to-end driving model deployed on Qualcomm SA8775P (QNN/HTP): deployment pipeline, INT8 quantization, fused custom op
-
Updated
Aug 28, 2026 - Python
SparseDriveV2 end-to-end driving model deployed on Qualcomm SA8775P (QNN/HTP): deployment pipeline, INT8 quantization, fused custom op
Closed-loop CARLA + Apollo integration of TensorRT end-to-end driving models (VAD, SparseDriveV2): model-agnostic relay, host-side TRT services, staged gates
End-to-end TensorRT deployment of SparseDriveV2 (3D detection + mapping + motion planning) on Jetson AGX Orin and x86: TRT plugin, ONNX export, FP16/INT8 engines, accuracy gates, latency/power bench
DualOS (Linux GPU inference + QNX 8.0 KVM guest deterministic scheduling) runtime research on AGX Orin — research-only, non-commercial
To associate your repository with the sparsedrive topic, visit your repo's landing page and select "manage topics."