Autonomous Decision Making Process TRIZ Cases

Biased scenario sampling raises rare adverse events in autonomous vehicle simulations, then correction factors recover accurate validation metrics.

Fusing machine-learning and model-based trajectory estimates improves long-term prediction reliability while capturing vehicle interactions.

V2V path sharing lets a following vehicle copy the lead trajectory, steer automatically, and keep a safe gap in complex convoy routes.

AI-based route planning uses battery state, power demand, and driving conditions to extend autonomous vehicle range without charging.

Delivery schedule alerts guide the driver to park where storage-door clearance is available, improving unattended parcel drop-off reliability.

Embedded emitter strips send electromagnetic lane and position data to vehicles, maintaining guidance when weather or obscured markings limit sensors.

An interface control unit routes automatic and manual driving signals to cut in-vehicle wiring complexity and connection cost.

Independent memory logging preserves autonomous vehicle sensor data during power loss, supporting crash analysis and configuration updates.

Lead-vehicle follow guidance helps autonomous cargo trucks switch modes near depots and park safely without adding trailer sensors.

Lead-follow driving and remote assistance help autonomous cargo trucks switch modes on surface streets and park at tight facilities.

Autonomous storage positioning lets traveling vehicles overlap or stand in parallel, cutting storage space and manual handling in facilities.

Probability-based surprise assessment helps autonomous vehicles choose trajectories that road users can better anticipate, reducing unsafe reactions.

Environmental sensing and server alerts let an autonomous vehicle halt, reroute, or evacuate users when floods, earthquakes, or tsunamis threaten.

Differential object tracking prioritizes nearby driving-relevant agents to cut compute load, heat, and storage use in autonomous vehicles.

Candidate speed profiles are smoothed and optimized in real time to handle changing limits, cut mechanical stress, and improve autonomous driving safety.

Direct belief state computation from current vehicle sensor data limits error buildup and supports reliable real-time control decisions.

Integrated waveguides and splitters replace bulky fiber coupling in coherent LiDAR, enabling compact multi-channel IQ detection with lower interference.

Replay-based simulation compares updated and baseline autonomous vehicle controllers to catch unintended behavior while limiting compute load.

A trigger-oriented rotating vehicle camera replaces multiple fixed cameras to maintain 360° coverage, redundant sensing, and lower power use.

Different display modes flag when position and region data use different reference stations, helping operators choose autonomous travel regions accurately.

Hazard recognition lets an autonomous vehicle halt, reroute, or evacuate based on earthquakes, floods, tsunamis, and heavy rain.

Multiple communication interfaces link mobile devices, servers, and the vehicle bus to enable keyless access, remote diagnostics, and updates.

Partial sensor, compute, and power redundancy preserves minimum field of view and control so an autonomous vehicle can reroute, slow, or stop after failures.

When camera, radar, or LiDAR detection fails, the controller switches fusion tracks by sensor reliability to maintain braking and deceleration control.

Detects machine health and job quality issues, filters false positives, and routes documented exceptions to supervisors for safe action.

V2X data elements for braking distance, autonomy status, and maneuverability help vehicles plan intersection access, spacing, and lane changes.

Fleet signals and sensor data identify aisle obstacles, then assign a suitable vehicle to clear them and keep warehouse traffic moving.

A dual-module neural architecture builds a virtual self-model to recognize agency, match temporal patterns, and predict actions in dynamic environments.

Linked parent-child accounts let autonomous vehicles enable child-specific access, notifications, and ride controls for safer, more comfortable trips.

Sensor-based ML generates vehicle motion segments from constraints and environment features to avoid obstacles with fewer violations.

Graphical matching of actual and lever-based speed and trajectory enables safer transfer from automated to manual vehicle control.

A separable autonomous chassis swaps vending machine body units at scheduled maintenance times, cutting remote service trips and operating costs.

Buffer-based policy selection lets autonomous vehicles balance safety, social acceptability, and passenger preferences in scenario-specific control.

Mapped traffic light regions and configuration-specific classifiers improve state detection accuracy under varied layouts and suboptimal images.

A mixed-effects and kriging model infers event risk at sparse locations, helping autonomous vehicles choose safer routes with limited data.

Checks autonomous vehicle motion plans against collision, traffic, and vehicle constraints before execution to prevent unsafe trajectories.

Emergency autonomous driving adapts vehicle maneuvers after sensor failure, then enables a safer stop or manual handover.

Short-term intention labels are combined with long-term outcomes to create better trajectory training data for more accurate AV prediction.

Interior sensing and user alerts help autonomous vehicles detect phones left behind at arrival and support timely retrieval before reuse.

Interior cameras detect occupant traits to auto-adjust seats, mirrors, and climate settings, reducing repeated setup in shared vehicles.

Sensor-guided yard tractors enforce speed and path rules to dock and move trailers safely in tight logistics yards.

Multi-sensor vehicle surveillance compares live and reference data to detect hazards and trigger actions even when the vehicle is turned off.

Virtual scenario primitives are merged with live sensor data to test autonomous vehicle responses in rare or unsafe real-world conditions.

Lane-based self-calibration corrects forward camera yaw offset for accurate lane centering despite installation variation and road disturbances.

A motorized sensor carrier rotates AV sensors to reduce blind spots while preserving cable routing and integrated sensor cleaning.

A cumulative trust score across time sources helps automotive systems resist drift and malicious time updates while preserving accurate system time.

Camera-based navigation checks jurisdiction-specific road features to avoid false turn-across-path alerts and guide remedial actions.

Correlating tire pressure and axle inclination improves pressure-loss detection and guides emergency braking for safer vehicle response.

When curb access is difficult from the opposite lane, the vehicle uses a same-side driveway to improve pickup safety and reduce passenger crossing.

Recoverable error handling gives the planner time to regenerate a trajectory while fallback control keeps an autonomous vehicle safe.