Autonomous Decision Making Process TRIZ Cases

A folded optical path with retro-reflection emulates long target distance, enabling accurate vehicle sensor calibration in compact service bays.

Overlapping radar, camera, ultrasonic, and Lidar sensing maintains environment observation after partial sensor failure for safer autonomous driving.

Multi-frame CNN analysis helps autonomous vehicles distinguish braking from turning signals across varied vehicle types and conditions.

A shared controller coordinates machines and unmanned aircraft to avoid part-related standstill, manage energy use, and reduce control complexity.

A tilted optical element creates distance-based light intensity so one 3D LiDAR sensor avoids near-field saturation and still detects far obstacles.

Indirect headway and speed control helps mixed-autonomy platoons stay stable when manually driven vehicles disrupt formation.

Underused vehicle control units process partial tasks from a central office, expanding computing capacity without new infrastructure or cloud reliance.

Dynamic vehicle repositioning matches formation roles to computing capability, improving vehicular micro cloud speed, accuracy, and resource use.

A TDC and multi-pixel photon counter replace costly high-speed ADCs to capture lidar peak time and intensity with wider dynamic range.

A motorized wheeled assembly turns a stationary directional fragmentation weapon into a remotely controlled platform for precise terrain-based engagement.

Extrinsic conditions reshape sensor observation area, letting the vehicle recalculate lane-change time from actual detection capability.

When primary sensors fail, secondary sensors guide the autonomous vehicle along a safe curvature to roadside parking while avoiding nearby obstacles.

Delta imaging and neural networks cut sensor and processing load while preserving accurate obstacle detection for autonomous vehicle control.

Distributed autonomous vehicles compare images and motion context to identify vehicles more accurately and track speed, direction, and trajectory.

Spatially coordinated vessel separation cancels Kelvin wake waves to cut drag without hull changes or mechanical links.

Private-key authentication secures vehicle-to-host updates, blocking tampering and preserving data integrity for autonomous operation.

A movable in-cabin command module lets one operator monitor multiple autonomous vehicles and respond faster to disturbances such as plugged seed tubes.

Shared obstacle data and AI-based verification help autonomous vehicles filter false positives before changing navigation paths.

Cross-training between camera and radar improves object classification, calibration, and distance-speed sensing for robust ADAS perception.

A gondola attachment lets autonomous vehicles cross roadways and geographic barriers between separated park areas without ground-only routing.

Annotated parking-area video streams sent over Li-Fi help autonomous vehicles detect obstacles beyond onboard sensor blind spots.

Automated shelf scans and back-room inventory checks trigger vehicle restocking only when space and stock conditions are met.

Time-division multiplexing shares photodetector channels in a coherent LiDAR transceiver, cutting fiber-coupling bulk while adding channels.

A parking control system repositions autonomous vehicles to remove non-required gaps and increase parking capacity without blocking departure.

Marginal-likelihood BNN training integrates out weights and adds PAC bounds to stabilize deep uncertainty-aware physical system control.

Graphical virtual walls replace infrared, ultrasonic, and magnetic barriers to improve navigation accuracy without extra hardware.

Multiple flight control components share sensor inputs and actuator commands to avoid common-mode failures and keep aircraft control fault tolerant.

Cable-pull sensing lets a robotic pool cleaner detect tug direction and adjust propulsion for wall climbing, shape changes, and simple remote control.

Constraint-based waypoint planning helps aircraft generate runway approach paths that improve landing efficiency while maintaining safe navigation.

Curved arrays of low-cost LIDAR and infrared sensors detect nearby objects and trigger speed reduction or stopping on power equipment.

A user-confirmed boundary check verifies the mower's digital work map before operation, avoiding unsafe drift without high-cost positioning.

Variable trim-rate control uses vessel speed, pitch, acceleration, and engine speed to enable fast low-speed trim and safer high-speed adjustment.

Route frequency analysis and sensor-map location checks flag compromised autonomous vehicle control requests before unauthorized navigation spreads.

Weak reflected light is amplified and ambient-light interference is filtered to improve robot ground detection on low-reflectivity surfaces.

Overlapping ultrasonic sensors detect obstacle distance and location without collision, helping autonomous lawn mowers avoid damage and adapt to more conditions.

Track-node and vehicle log comparison enables objective autonomous vehicle testing, including reaction time and transition smoothness.

Autonomous aircraft use onboard sensing, landing warnings, and safety checks to deliver medical services with less operator dependence.

A sensor-equipped field rover adjusts sprinkler water by location and elevation to prevent frost damage while limiting water waste.

User-reported cleanliness or damage triggers remote dispatch actions, helping autonomous service vehicles stay suitable without constant oversight.

Relevant sensor feeds and tools are selected for AV exceptions so tele-operators can resolve obstructions faster with less intervention.

Multiscale potential surfaces and sensor feedback help autonomous vehicles avoid local extrema, cut turns, and fully cover unknown spaces.

When bad weather blocks a drone route, this case shows how recipient-selected destinations and surface transport cut delivery delays.

Local inference in stacked sensor memory cuts raw image traffic to the host, easing CPU load while supporting higher frame rates.

AHaH feature extraction compresses noisy LIDAR point data into stable low-dimensional features, cutting power use for real-time autonomous driving.

A machine-learned gaze-based ROI lets vehicle control units classify relevant objects without processing all sensor data equally.

Sensor-driven online learning updates control signals and tool paths so earth moving vehicles adapt to changing soil and vehicle conditions.

Beacon mapping aligned to geographic reference points replaces boundary wires and supports GPS-guided mowing paths with simpler setup.

When a guided vehicle fails to switch paths cleanly, heading comparison and sensor feedback establish a corrected route before drift causes damage.

When terrain or nearby objects block sensor views, the vehicle shifts to a stop location that preserves unobstructed 360-degree threat monitoring.

Sensors and path replanning let a self-driving vehicle slow and reroute around obstacles, avoiding emergency stops and manual restart.