Emergency braking systems (EBSs) are critical for maintaining vehicle stability and safety during severe braking maneuvers; however, conventional EBS controllers often rely on wheel-slip feedback and exhibit conservative braking performance due to the highly nonlinear and time-varying nature of tire–road interactions.To address these limitations, this paper proposes a direct torque-level emergency braking framework for electric vehicles that maps readily available onboard measurements directly to four-wheel braking torques without requiring explicit estimation of wheel slip, tire–road friction, or vehicle sideslip states. An expert policy is first constructed using residual reinforcement learning, in which a modified twin delayed deep deterministic policy gradient variant (TD7) agent learns corrective braking actions on top of a model-based controller. The resulting expert policy is subsequently distilled into a lightweight student controller through a teacher–student learning framework. The proposed approach is evaluated in a high-fidelity vehicle dynamics environment and further validated through cross-platform verification. Experimental results demonstrate superior braking performance compared with a conventional MPC controller in terms of stopping distance, braking time, and vehicle stability under diverse road-friction conditions. The results further confirm the robustness and cross-platform generalization capability of the proposed framework, highlighting its potential as an effective and computationally efficient solution for intelligent emergency braking control in future electric vehicles.

Mapless navigation often removes global maps while retaining localization-derived goal vectors or bearings. We study a stricter setting in which a mobile robot observes only local LiDAR, scalar goal range, and short histories of executed actions; neither pose nor goal direction is provided to the policy. We introduce ACR-Nav, an action-conditioned range navigation framework that converts scalar-range evolution into closed-loop progress information. Its range–action history associates each distance change with the motion that produced it, while the sectorized LiDAR captures local geometry and short-term obstacle motion. A LiDAR-only safety filter provides immediate collision intervention, and a static-to-mixed curriculum stabilizes learning. A lightweight multilayer–perceptron is optimized with Proximal Policy Optimization (PPO), while the ACR-Nav formulation itself remains optimizer-agnostic. In corridor simulations, ACR-Nav achieved 93.2%, 80.4%, and 84.4% success in static, mixed, and dynamic environments. Removing the safety filter reduced success by 15.2, 14.6, and 16.0 percentage points in static, mixed, and dynamic environments, respectively, and random-goal tests yielded 91.2% and 81.4% success in static and mixed settings. Topology-shift experiments further quantified adaptation to an L-shaped corridor. The results show that action-conditioned scalar-range evolution can support goal-directed, segment-level navigation within locally straight corridor passages without exposing robot pose or target bearing to the policy.