External Condition Input Parameters TRIZ Cases

A common MPC envelope uses human and pseudo-driver inputs to unify semi- and fully autonomous control while reducing hardware.

Neural networks turn nearby objects into virtual elastic and inelastic fields, improving autonomous driving robustness under noisy, occluded sensing.

Tracks generation time and data age across processing blocks so distributed systems can detect latency breaches and avoid outdated responses.

Driver-specific limit ranges let assistance switch to brief autonomous control only in critical situations, improving safety without unnecessary intervention.

Highlighted test-related driving evaluation items help drivers quickly see test eligibility without losing full diagnosis feedback.

External conditions and vehicle data are combined to vary platoon following distance, preserving fuel savings while maintaining safe spacing.

Procedural scenarios use object trajectories and intersection probabilities to verify vehicle safety predictions across complex driving cases.

Driver steering input is prioritized during autonomous lane changes, with stop-condition checks to continue control or announce handover.

Dynamic risk potential fields shift the steering path toward the vehicle position to avoid front objects without excessive steering or deceleration.

Sensor fusion compares a color slope map with a simulated bubble to warn drivers of wrong-way travel without confusing roadside signage.

A coordinating controller merges ACC and predictive cruise commands by choosing the lower acceleration or higher deceleration to protect safety and fuel economy.

Monitored sensor and actuator degradation is used to retune planning and control parameters, helping automated vehicles avoid unsafe maneuvers.

A 3D lane display shows road curvature ahead from sensor data, helping drivers read curves quickly and react safely.

Learns driver lane-position preferences from behavior and road context to shift vehicle placement within a lane for greater comfort.

Detects adjacent-lane gaps and adjusts following distance and speed so vehicles can merge smoothly in dense traffic with lower collision risk.

Risk-weighted inputs tune ADAS warning thresholds and incident alerts to driver behavior, vehicle state, and road hazards.

Wireless state data from nearby motorcycles helps cruise control handle curves more safely when onboard information is insufficient.

Object data is transformed at a trajectory reference point into tangential coordinates, cutting path-analysis complexity for hazard detection.

Actuator-delay-aware steering and torque vectoring help vehicles track an emergency avoidance path while maintaining stability and tire limits.

After an abnormal driver state is detected, the vehicle stops first, then grants limited remote control to reduce hypothermia risk and misuse.

When level 3 conditions are unmet, level 2 speed and lane alignment control prepares the vehicle state and reports the adjustments to the user.

Real images enriched with simulated road debris expand scarce AV training data, improving object detection and hazard assessment.

Dynamic safety-distance and jerk control helps ACC handle cut-in vehicles while reducing unnecessary braking and preserving ride comfort.

Nonlinear predictive control uses grade profiles and headway data to cut platooning fuel use on steep terrain while avoiding excess braking.

Historical path features guide autonomous driving policy learning, cutting retraining time, data demand, and vehicle wear for new tasks.

Vehicle download control uses position-based failure feedback to avoid poor communication areas and improve update success.

Validates planned driving paths by checking lateral range, motion direction, and acceleration to block unsafe maneuvers without rejecting defensive responses.

A speed-adaptive protruding detection region improves cut-in vehicle recognition at low speed and when lane lines are unclear.

A trajectory-fitted warped occupancy grid cuts storage and processing on curved roads, enabling faster object detection in an autonomous vehicle's path.

A timed deceleration plan lets an autonomous vehicle pass over road objects while accounting for following-vehicle reaction time and damage risk.

Numerical integration with a bicycle model improves yaw rate and heading references for low-speed autonomous trajectory tracking.

Neural networks compare learned and actual steering to correct curved-road disturbances and adjust speed for more stable lane following.

AI detects hearing-related driving conditions, then triggers sensory aids and vehicle motion changes to help drivers notice critical road cues.

A predictive road elevation profile uses sensor, location, and map data to improve vehicle control through bank and grade transitions.

Fixed roadside sensors send object position and trajectory data to approaching vehicles, extending detection at intersections and curvy roads.

Predicts sneeze onset and phases to adjust steering and braking control, helping maintain vehicle stability during driver incapacitation.

An asymmetric assist boundary shortens unnecessary collision-avoidance control when crossing targets move away, improving occupant comfort.

Rule-based nudge planning uses lane layout, vehicle width, and speed to route around obstacles when enough lateral space is available.

Zone-based user position detection triggers vehicle pickup at the right time, cutting autonomous valet wait time across parking floors.

CW radar filters collision-relevant transverse motion from velocity and angle trends, then FMCW activates for precise tracking with fewer false reports.

Rear-biased braking uses road friction and motion-state sensing to maintain vehicle attitude and prevent lane departure on slippery roads.

Blockchain-based checks verify vehicle components and operator association, then selectively limit unsafe functions when authorization fails.

Driver-specific following behavior is learned with Gaussian Process regression, then safety-filtered to deliver personalized ACC spacing and acceleration.

Camera-guided steering limits cap lateral acceleration to keep autonomous navigation safe in complex environments while reducing steering jerk.