A mixed controller coordinates steering and differential braking to evade obstacles quickly while keeping the vehicle controllable for smooth driver takeover.
A unified hybrid power gauge combines electric and combustion power into one moving indicator to cut key presses and reduce driver cognitive load.
Switching acceleration limit modes during wheel slip improves estimated vehicle speed accuracy despite pitching-induced sensor deviation.
When a crosswalk is detected, the assistance zone expands outward to cover the full crossing and reduce pedestrian detection delay.
Switching from distance control to speed control at higher lead-vehicle speeds cuts repeated coasting and powering changes to save energy.
Alignment modules bridge human-defined and AI-defined interfaces to improve task coordination, interpretability, and update flexibility in automated driving.
When abnormalities occur in autonomous driving, the vehicle pre-identifies shoulder candidate zones and selects a minimal-risk stop position.
When an oncoming vehicle may pass a slow mover on a narrow road, control thresholds are relaxed to trigger earlier braking and cut contact risk.
Adaptive assistance thresholds and intervention tuning help establish automated driving conditions with less driver distraction and smoother autonomy handoff.
Predefined behavior modes let an automated vehicle switch around failed modules and keep control within safe boundary conditions.
Driver-set torque reduction is executed automatically during gear shifts to cut wheel slip and improve launch repeatability.
Using left and right side velocity data, this case improves low-speed vehicle MPC accuracy and steering precision without relying on overall vehicle speed.
Motion, driving pattern, and cabin sound data are fused with ML models to identify vehicle type and user role for continuous driving monitoring.
A mode-selectable paddle switch lets the ECU switch between shift and speed control, reducing driver workload when road conditions change.
Redundant automated driving control waits before handover to manual mode, allowing fault-based switching and avoiding premature control loss.
By restricting feedback control when actuators cannot respond, this vehicle controller avoids abrupt acceleration changes and preserves ride comfort.
Machine-learned divergence weighting helps autonomous vehicle trajectory validation separate material from immaterial deviations at scale.
Wheel speed feedback lowers the engine slip target under split-friction conditions to balance driven wheels and preserve tractive force.
Transfers driving authority only when road conditions and driver skill indicate a smoother manual takeover with lower risk of abrupt braking or turning.
A centralized arbitration and profile generation layer unifies multi-feature motion requests to cut duplicate torque conversions and improve drive quality.
Lateral warning zones, heading-angle checks, and TTC thresholds help detect blind-spot targets crossing into the ego-vehicle path.
Radar road-edge sensing predicts lane position when camera markings disappear, helping maintain accurate vehicle control.
A two-stage radial and Bezier-based non-radial correction flow flattens residual shading and improves wide-angle image stitching.