Autonomous Decision Making Process TRIZ Cases

Just-in-time AGV assignment, battery-aware charging, and adaptive path planning cut wasted trips and avoid mid-task recharging delays.

Dual monitoring devices cross-check vehicle component faults in real time, improving alarm accuracy and reducing false alerts in autonomous driving.

When camera or map data fails, radar-detected objects and arc-curve selection provide a fallback reference line for autonomous vehicle trajectory planning.

Mirrors redirect onboard sensor beams around vehicle obstructions to cut blind spots without adding sensors or processing load.

Detects a mobile device left in an autonomous vehicle after arrival, then alerts the user and supports nearby vehicle repositioning for retrieval.

Hysteresis-based user following improves ergonomic comfort and pedestrian courtesy while enabling autonomous doorway passage and convoy travel.

A learned yield model replaces hand-coded driving rules to make faster, scalable right-of-way decisions from sensor and map data.

Real-time surroundings on a touch screen let riders refine drop-off points while the vehicle slows or pauses at stoppable locations.

A detachable autonomous driving kit lets standard vehicles self-transport through cooperative control, cutting manpower, delays, and damage risk.

An ANN-based prediction and update estimator improves ego-vehicle state accuracy and robustness despite noisy sensors and changing conditions.

Target and capability maps reveal where vehicle sensors fall short for a task, guiding repositioning or parameter changes to maintain coverage.

Visual passenger identifiers let autonomous vehicles confirm rider accounts and refine pickup points in crowded areas without app-based hailing.

Shows only consequential nearby objects and route adjustments on the vehicle interface, reducing clutter and strengthening passenger confidence.

Mobile charging vehicles reorder EV queues, schedule rendezvous charging, and route themselves to reduce range anxiety and waiting time.

Driving parameters are adjusted from nearby vehicles' autonomy metrics to balance separation, speed, safety, and traffic flow.

Sensor-detected passenger identifiers help autonomous vehicles verify and locate riders in crowded areas without app or network dependence.

By adding hyperspectral sensing to LiDAR, this case separates solid obstacles from exhaust plumes and vapor to avoid unnecessary maneuvers.

When conservative AV rules cause merge deadlocks, passenger-driver negotiation creates safe gaps that cut waiting time and ease traffic flow.

Real-time vehicle signals and LiDAR obstacle data let a warehouse fleet assign clearing tasks, reducing aisle congestion and collisions.

Separate steering and speed command modules use trajectories and constraints to improve autonomous navigation accuracy in dynamic environments.

Predefined road-region templates adjust vehicle position and dynamics to improve highway traffic flow while reducing coordination burden and collisions.

Adaptive point density and weight tuning improve autonomous vehicle trajectory safety, comfort, and computing efficiency.

By delaying the next HMI instruction until the prior response is done or predicted complete, the vehicle reduces driver overload and improves response accuracy.

Authenticated external triggers let emergency vehicles command autonomous vehicles securely, reducing missed detection and false positives.

Selectable screen areas let work vehicle operators show or hide travel-state data, reducing clutter while keeping location guidance available.

External control signals guide a self-propelled roll trolley to handle varying roll sizes, avoid bottlenecks, and move safely without fixed rails.

Predictive dispatch uses rider behavior, battery status, and connectivity to avoid missed autonomous vehicle pickups during device disconnects.

Dynamic interflow judgment shifts handover notification timing and deceleration to preserve automatic lane merging while maintaining running safety.

AR carpet images mark current and target lanes with real-time vehicle data, improving lane-change accuracy and collision prevention.

Pretrained user profiles and scenario-choice data let autonomous vehicles match passenger driving preferences, even in emergency conflicts.

Independent auxiliary control units process sensor data and trigger steering or braking responses when the primary autonomous driving controller fails.

When map-based routing fails, the vehicle learns shared road trajectories from nearby traffic to generate a usable travel strategy.

Media analysis drives seat, lighting, climate, and motion control in autonomous EVs to create a more immersive in-cabin experience.

Photorealistic 4D traffic scenes blend real and simulated data to improve autonomous driving model training realism without field-test cost.

Modular delivery robots combine shared mapping, localization, and fleet coordination to handle diverse transport tasks with less redesign.

Piecewise linear chirps let one FMCW lidar deliver fine near-range resolution and coarser long-range sensing without multiple sensors.

Probabilistic collision zones combine trajectories and crosswalk context to cut unnecessary yielding while protecting pedestrians.

Machine learning predicts pedestrian and bicyclist intent from hidden context to generate more realistic autonomous vehicle test scenarios.

Environmental anomaly checks let a vehicle control unit detect off-vehicle attacks and switch to a reduced-function mode to block misuse.

When a location marker is missed, the transport vehicle records its motion, retraces to a known point, and retries the route without human help.

Continued deceleration after manual takeover keeps braking consistent, reducing driver discomfort and stopping-distance risk.

When road views are blocked, the AV infers hidden object motion in occluded regions and adjusts speed and path to avoid collisions.

Authorization-gated state changes route autonomous vehicles through standby instead of direct manual-autonomous switching, improving transition safety.