top of page
Real-time VLM Optimization for E2E Autonomous Driving
- Vision-Language Model
- End-to-End Framework
- Model Quantization
System & Platform, Data-driven AD
2026
- TensorRT Optimization for VLM-based E2E Framework
- Optimizing the VLM-based E2E model to meet real-time driving constraints.
- Practical validation and hardware feasibility study
- Validating deployment feasibility via TensorRT engine conversion and quantization (FP16/INT8).
- Benchmarking inference latency to ensure reliable decision-making on edge devices.
bottom of page