https://xuquant.com/posts/autodrive/Recent content in Autonomous Driving: End-to-End, VLA, and Beyond on Xu'Bloghttps://xuquant.com/images/profile.jpghttps://xuquant.com/images/profile.jpgHugo -- 0.152.2enWed, 29 Apr 2026 14:00:00 +0800https://xuquant.com/posts/autodrive/qwen3-vs-qwen3-5-architecture/Wed, 29 Apr 2026 14:00:00 +0800https://xuquant.com/posts/autodrive/qwen3-vs-qwen3-5-architecture/A deep architectural comparison of Qwen3.5 versus Qwen3, examining hybrid attention, native multimodal fusion, high-sparsity MoE, and partial RoPE across attention, vision, and MoE dimensionshttps://xuquant.com/posts/autodrive/basic_rl/Tue, 20 Jan 2026 10:00:00 +0800https://xuquant.com/posts/autodrive/basic_rl/Comprehensive analysis of applying reinforcement learning to end-to-end autonomous driving, covering metric caching, Direct Preference Optimization (DPO) across action representations, and strategies for breaking sampling ceilings in iterative self-improvement.https://xuquant.com/posts/autodrive/nvidia_vla/Sun, 11 Jan 2026 10:00:00 +0800https://xuquant.com/posts/autodrive/nvidia_vla/Technical deep-dive into Nvidia's Cosmos-Reason (Alpamayo) VLA system for autonomous driving, covering tri-plane vision encoding, ego-shortcut avoidance, Cause-of-Change dataset paradigm, and reasoning-action alignment via reinforcement learning.https://xuquant.com/posts/autodrive/e2e-autonomous-driving-evolution/Thu, 01 May 2025 10:00:00 +0800https://xuquant.com/posts/autodrive/e2e-autonomous-driving-evolution/A technical survey on the architectural evolution of end-to-end autonomous driving, covering planner decoder selection (AR vs Diffusion vs Flow Matching), VLA integration strategies, and engineering best practices for data infrastructure, training optimization, and evaluation systems.https://xuquant.com/posts/autodrive/rl-policy-optimization-e2e-driving/Wed, 30 Apr 2025 10:00:00 +0800https://xuquant.com/posts/autodrive/rl-policy-optimization-e2e-driving/A systematic derivation of policy optimization methods for end-to-end autonomous driving: from REINFORCE through PPO to GRPO, covering advantage estimation, sampling differences between LLM and driving, multi-objective loss design, and the role of noise in diffusion-based exploration.https://xuquant.com/posts/autodrive/inspatio-world-4d-simulator/Sun, 20 Apr 2025 10:00:00 +0800https://xuquant.com/posts/autodrive/inspatio-world-4d-simulator/A deep technical analysis of InSpatio-World: a 1.3B-parameter real-time 4D world simulator that combines implicit spatiotemporal caching with explicit geometric constraints, achieving 24 FPS novel-view synthesis from monocular video.https://xuquant.com/posts/autodrive/ar-trajectory-tokenization/Tue, 01 Apr 2025 10:00:00 +0800https://xuquant.com/posts/autodrive/ar-trajectory-tokenization/A deep dive into trajectory tokenization for autoregressive driving planners: from state-based discretization via k-means clustering, through token matching and reconstruction, to the AR+Diffusion paradigm and GRPO-based reinforcement learning post-training.https://xuquant.com/posts/autodrive/generative-planning-nonconvex/Sat, 15 Mar 2025 10:00:00 +0800https://xuquant.com/posts/autodrive/generative-planning-nonconvex/A first-principles analysis of why regression-based planners fail in autonomous driving: the feasible set is non-convex, MSE averages into obstacles, GMM is a patch not a solution, and generative approaches are necessary.https://xuquant.com/posts/autodrive/deepseek_series1_mla/Sat, 15 Feb 2025 10:00:00 +0800https://xuquant.com/posts/autodrive/deepseek_series1_mla/Deep technical analysis of Multi-Head Latent Attention (MLA) from DeepSeek-V2, covering low-rank KV cache compression, decoupled RoPE design, and computational cost comparison with MHA, MQA, and GQA.