Autonomous Decision Making Process TRIZ Cases

Multi-sensor road sensing lets a mobile robot detect hazardous vehicles mid-crossing, then brake, reverse, or alert drivers to avoid collisions.

Street-mounted sensors share signed object data at intersections to extend vehicle detection beyond line of sight and improve navigation safety.

User inputs add directional, velocity, or acceleration changes to autonomous UAV flight paths without interrupting ongoing flight.

Semantic branches added to selected neural network layers turn black-box vehicle predictions into user-readable explanations with measured accuracy.

AI object tracking and sensor fusion help subsea vehicles navigate and perform tasks with less operator control despite low-bandwidth links.

Threshold-based input switching lets work vehicles exit autonomous travel only on intentional operator actions, reducing unintended manual takeover.

Dual user interfaces let operators modify UAV flight paths during autonomous missions for obstacle avoidance without full manual takeover.

Operator-tool change thresholds replace steering-state triggers to prevent unintended manual takeover during autonomous work vehicle travel.

Multiple modems and network map data let a vehicle route sensor and vehicle data around congestion to maintain reliable remote communication.

Multi-range sensors feed an occupancy grid that identifies discontinuous surfaces and triggers real-time vehicle reconfiguration for safe traversal.

Operator profiles built from past assisted autonomy tasks help match remote vehicle assistance requests faster with less bandwidth and downtime.

An event-triggered T-S fuzzy H∞ controller helps unmanned surface vehicles stay stable under DoS attacks while reducing network traffic.

Motion control turns internal data transmission and reception into visible robot actions, improving user awareness and engagement in real time.

Resetting stored drive data when picking begins helps semi-automated forklifts adapt acceleration without using outdated manual driving behavior.

Offset corner thrusters steer exhaust away from the seabed and camera, reducing silt disturbance while preserving precise underwater control.

Cryptographic checks on ANN partitions verify stored weights and biases before use, helping detect integrity faults in autonomous systems.

A dual-assurance architecture validates autonomous flight commands from unstructured sensor inputs, enabling certifiable aircraft control.

Trip-based turn priority bids let autonomous vehicles resolve occluded intersections with less motion-planning computation and smoother flow.

Clearance-driven route graphs let unmanned aircraft plan executable taxi paths that match kinematics, avoid collisions, and cut manual planning effort.

Sensor-based human detection lets an autonomous mobile robot stop, reorient carried items, or avoid obstacles while limiting noise in sensitive areas.

Onboard LiDAR, fisheye cameras, radar, and AI enable autonomous navigation and 3D mapping when wireless links or external data are unavailable.

Time-based sharing of connection IDs lets multiple mobile bodies receive positioning correction data without wasting IDs or raising operator cost.

Separating self-reflected and external ToF light reduces cross-talk noise, improving depth maps and motion guidance for autonomous vehicles.

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

A distributed ledger logs autonomous vehicle events to enforce liability smart contracts and accelerate objective claims handling.

Map and localization data guide multi-camera exposure so critical road regions stay visible despite sudden lighting changes.

Sparse LiDAR points are grouped into vertical pillars and converted to pseudo-images so 2D CNNs can detect 3D vehicle surroundings in real time.

When onboard driving logic stalls, a control center uses live sensor data to guide an autonomous vehicle to an intermediate position.

Shared sensor data from nearby vehicles and fixed sources fills blind spots and improves object-position mapping for autonomous navigation.

A multi-objective planner balances observation points with thermal or wave energy harvesting to extend ASV endurance without losing mission time.

GNSS and IMU tracking with Lurie-Postnikov attraction domains helps detect AMR instability early and trigger safe path termination.

Adaptive UAV position reporting raises update frequency when collision risk increases, improving avoidance while limiting energy use and signaling load.

Real-time cloud sensing and machine learning guide unmanned seeding vehicles to target suitable clouds and reduce manned flight cost and risk.

Sensor and repair-history analysis forecasts repair outlays, groups work for quantity savings, and cuts vehicle downtime and labor strain.

Stopping at inter-field route points and announcing travel status helps autonomous work vehicles avoid road conflicts and resume safely.

Real-time positional and kinematic data keep an autonomous grain cart aligned with a harvester, reducing misalignment, delay, and collision risk.

Weighted scene metrics replace equal-mile testing to assess autonomous system readiness with higher confidence and less driving data.

Marker recognition with preplanned action control helps an autonomous mobile robot execute user-intended motions quickly and reliably.

An aerial assist drone fills occluded areas and supplements AV sensors to improve navigation reliability without adding onboard sensor complexity.

Floor models and 3D tunnel maps enable accurate underground mobile object positioning without added altitude sensors or full 3D dead reckoning.

Coordinated drone flight paths and camera aiming maintain formation around moving subjects in real time, enabling complex multi-angle filming.

Visible light links and optical random numbers let autonomous vehicles assign right-of-way directly, cutting RF latency and relay dependence.

Cross-checking detected vehicles or pedestrians against road regions helps judge map reliability and switch automated driving modes safely.

Location signals from autonomous vehicles verify real visits, enrich user reviews, and improve destination recommendations.

Camera-based marker detection and virtual boundary control keep an autonomous work machine inside the working area when markers are obscured or move.

An upward-facing RFID reader limits stray shelf tag reads, improving mobile item identification and on-site shipping label printing.

Real road meshes, network data, and supplemental tertiary data are combined to build scalable simulated worlds for autonomous vehicle training.

GNSS, terrain sensing, and onboard control let working vehicles slow for sharp curves and rough zones before damage while staying autonomous.

On-board sensors and AI detect obstacle-caused stops, then create virtual stop lines to map unmarked road features and improve navigation safety.

Sunshine condition data is used to tune robot sensor parameters, reducing detection variation and improving autonomous navigation accuracy.