Autonomous Decision Making Process TRIZ Cases

Feature map dependency loss regularizes neural network training to curb overfitting and use unlabeled data for better classifier accuracy.

A retractable wheel lets an autonomous robot stand while moving and sit while stopped, easing attitude control and cutting idle power use.

Pixel-to-line distance and angle encoding preserves spatial detail through down-sampling, improving real-time line detection for autonomous driving.

Separate coherent intervals for Doppler shift and range correction improve velocity sign detection and reduce phase-related range noise.

Environmental modeling and coverage estimation let waterborne vehicles reposition autonomously to maximize target search range with less manual control.

Stored GPS and unloading-point data let an agricultural vehicle return a receiving container to the last fill position, reducing spillage and manual input.

Sensor and sub-system data are used to assess damage, notify the right parties, and shorten autonomous vehicle recertification.

Peer-to-peer blockchain verification lets ADS-equipped vehicles validate ODD-specific features with lower infrastructure cost and traceable results.

Dynamic cart assignment and picker guidance reduce travel time, balance workloads, and improve resource use in single-line order picking.

Terrain-aware path spacing in slope areas keeps overlap within range, preventing missed coverage and improving full-coverage work.

Temporal analysis of paired railroad light signals helps autonomous vehicles cut false detections and adjust approach behavior by confidence.

Shortest-path routing is varied with random intra-regional points so autonomous mowers avoid repeated tracks and lawn ruts.

By modeling many users as one crowd, the robot plans movement and guidance with less confusion and smoother navigation in public spaces.

Multi-agent reinforcement learning coordinates drone taxi routes and passenger selection to improve response time, cooperation, and profit.

Anticipatory crop row maps combine aerial sensing with real-time updates to guide autonomous farm robots through mis-planted rows, weeds, and obstacles.

Localization-based slope compensation converts gravity-induced forces into wheel torques to keep autonomous lawn mowers on straight paths on slopes.

An above-ground self-propelled irrigator uses sensor-driven nozzle and flow adjustment to cut water waste and avoid buried sprinkler upkeep.

Sensors verify pile location and orientation against tolerance thresholds so autonomous pile driving can trigger corrective action and avoid rework.

Real-time collision location and timing prediction helps mobile robots cross roads safely around moving vehicles and pedestrians.

Time-based image analysis of paired railroad lights lets autonomous vehicles estimate activation confidence and avoid false crossing decisions.

Asynchronous fusion of lidar, camera, radar, and ultrasonic data keeps occupancy grid maps accurate, timely, and usable when one sensor fails.

Real-time stereo camera and LiDAR mapping replaces external HD maps, enabling precise autonomous navigation in GPS-poor or indoor settings.

Two AI models separate visible and partially blocked object detection, helping mobile electronics avoid collisions in obstructed paths.

Coordinated robotic refuse containers share battery, location, and fill status to trigger autonomous replacement and keep collection running.

Independently actuated wheel sets use sensor-guided control to shrink turning radius and maintain clearance for large autonomous vehicles in tight spaces.

Autonomous routing combines great-circle arcs, rhumb lines, and hard or soft points to avoid terrain and moving no-fly zones.

3D RF signal maps let a UAV reroute around high-strength zones during transmitter inspection, reducing interference damage and operator risk.

Blockchain smart contracts use vehicle event and control-state data to automate liability assignment and speed insurance claims.

Forecast and real-time weather data guide ML-based battery and path adjustments to keep automated vehicle missions on track.

Targeted sensor recording tasks let mobile units capture only weak driving cases, speeding control model updates and reducing data collection effort.

Outer-periphery obstacle detection shifts a work vehicle inward in headlands, preserving ridge-side coverage while avoiding contact.

Coordinated vehicle and port controllers automate checkpoint passage and loading handoffs, cutting driver dependence, cost, and fatigue risk.

Order-slot routing at network nodes cuts replanning complexity, synchronizes robot paths, and avoids collisions in confined spaces.

A blockchain ledger routes vehicle events to smart contracts to assign liability automatically across manual and autonomous control changes.

Randomized intra-regional routing helps autonomous mowing robots avoid repeated tracks while keeping travel between start and end points efficient.

Real-time sensor feedback updates machine learning control for earth moving vehicles as soil conditions and vehicle responses change.

Third-party communications and sensor analysis identify autonomous vehicle status remotely, enabling fast control actions such as stops or recovery plans.

Maintains occupied-region maps and connection scores to handle unseen obstacles outside sensor view, improving mobile apparatus navigation.

Autonomous UAVs and ground stations shift delivery into 3D routes, easing road congestion while improving navigation and last-mile speed.

Machine-readable visual markers let vehicles determine location and orientation in tunnels or urban canyons without GPS or remote links.

Arc-arranged LiDAR and infrared sensors expand machine vision coverage for power equipment while lowering object detection cost and operator burden.

Sensor thresholds and remote severity analysis help autonomous vehicles detect road impacts, avoid false inspections, and route needed servicing.

A UPID-based rider service layer uses passenger metadata to restore personal interaction in autonomous ride-hail trips and improve loyalty.

Directional acoustic sensing steers a drone toward unrecognized sound sources, reducing visual hardware, power use, and low-visibility limits.

Selective dropping of LIDAR points or pulses eases channel congestion while preserving critical environmental data for vehicle navigation.

Dynamic routing guides pickers through vehicle-restricted warehouse zones and sets efficient rendezvous points to cut walking time.

Directional light-pattern signaling between vehicles avoids RF interference and noise, enabling secure, reliable data exchange for autonomous driving.

A synchronized control pipeline schedules sensor processing and actuator updates to cut latency and jitter in real-time flight control.

Projects LiDAR point clouds to 2D and fits an optimized circle to separate pedestrian points from nearby objects with lower complexity.

Automatically detects whether a remotely controlled vehicle should park or exit, enabling forward or reverse maneuvers with less user input.