External Condition Input Parameters TRIZ Cases

Camera and radar data guide adaptive avoidance paths around pedestrians and two-wheeled vehicles while preserving lane position and traffic flow.

Blind spot regions with graded hazard levels enable more precise right-side vehicle warnings using LiDAR-based participant detection.

Yaw rate change predicts turning paths, letting driver assistance estimate contact points and times with lower processing load.

Two-stage camera and LiDAR sensing helps autonomous vehicles detect shoulder objects earlier and adjust operation with higher confidence.

When sensor-detected traffic signals conflict with map data, the guidance function suppresses unsafe automation and alerts the driver.

Learns driver steering behavior on curved roads during non-control periods, then adjusts lateral position control to reduce discomfort and anxiety.

Kinematics-based comparison flags weak AV trajectory predictions and reweights those cases to improve training accuracy.

Traffic-aware cut-in control uses lane congestion and neighboring vehicle intent to adjust host vehicle speed and keep flow smoother.

Radar pattern recognition identifies train-like object rows and suppresses unnecessary braking when no rail crossing is detected.

Feedforward torque vectoring uses road bank angle and a vehicle model to curb heavy-duty vehicle drift without jerky steering response.

Reinforcement learning predicts future safety zones so vehicle spacing adapts to traffic and road conditions before rear-end risks escalate.

Automatic calibration aligns LiDAR position and angle with odometer motion data, cutting manual setup time and route deviation.

A moving, color-changing cursor turns rear vehicle distance into clear visual cues, helping drivers judge spacing in low visibility.

Weighted similarity scoring selects diverse robot trajectories from ML predictions, cutting planning runtime while covering hazardous scenarios.

Weighted similarity scoring narrows predicted robot trajectories to the most relevant set, reducing planning time while preserving safe control.

Reinforcement learning updates vehicle control from command-quality feedback to reduce trajectory error and improve autonomous driving precision.

During manual maneuvering, steer-by-wire assistance predicts driver intent and corrects trajectory deviations to avoid collisions and improve parking.

Machine learning captures cut-in and cut-out driving behavior to personalize ACC responses, improving comfort and reducing driver override.

Lidar point clouds and distance-ratio analysis help identify cut-in vehicles and direction with lower computational load and high accuracy.

Dynamic control parameter adaptation adjusts motorcycle riding automation to rider input, behavior, and conditions for smoother guidance.

Gradient-corrected map markings are compared with detected lane lines to improve vehicle lane decisions on sloped roads.

Trajectory characteristics drive real-time maneuvering control adaptation to balance path accuracy and occupant comfort without external maps or sensors.

Camera-radar distance fusion applies a correction coefficient to detect curbs and steps accurately, improving travel control without LiDAR.

Surrounding-traffic assessment guides an autonomous vehicle to a safe lane or stop point before manual takeover, reducing collision risk.

Surrounding-traffic checks guide an autonomous vehicle to a safe lane or stopping place before handover, reducing collision risk during manual takeover.

Dynamic torque profiles switch by stability margin and torque drop to limit oversteer or understeer while preserving energy recuperation.

Checks recent driver set-speed changes and speed margin before applying a new target, avoiding unnecessary automatic acceleration or deceleration.

Planned trajectory characteristics drive manoeuvring control selection, improving autonomous driving accuracy and comfort without extra map or sensor input.

When a vehicle cuts in too closely, weighted ACC acceleration blends distance and speed difference to keep safe gaps with smoother braking.

Behavior-based scheduling predicts low-activity periods in similar vehicles so software updates can run with less downtime and disruption.

Candidate paths, covariance-based polygons, and speed envelopes let autonomous vehicles predict moving-object occupancy with less data and compute.

On-board rule evaluation cuts telematics delay and keeps vehicle actions responsive when network links are slow or unavailable.

During curve assistance, acceleration input triggers suppressed acceleration instead of abrupt response, balancing driver intent with safe corner entry.

Dynamic controller profiles and tree search balance responsiveness with smooth, low-jerk autonomous vehicle maneuvers.

Risk-aware sampling steers autonomous vehicle motion planning toward rare dangerous agent behaviors, improving robustness with fewer simulations.

Combining engine response, LiDAR, and image data improves road grade prediction where GPS and IMU estimates lose accuracy in urban driving.

Sensor fusion and machine learning infer driver intent and surroundings to choose vehicle direction automatically and reduce mode-selection errors.

Consensus-validated hazard detection lets vehicles warn nearby devices and pedestrians using shared speed, direction, and distance data.

Real-time safety scoring guides discretionary lane changes, balancing driving convenience with complex environmental assessment.

A coupling-point path strategy splits consecutive curves into two segments to reduce steep turns and sudden deceleration for smoother driving assistance.

A processor checks driver input, ODD status, device health, and field of view before enabling discretionary lane changes.

Sensor-guided slipstream positioning cuts aerodynamic drag at higher speeds, lowering EV energy use while extending driving range.

Predicts adjacent vehicle trajectories and warns riders when lane-splitting gaps are likely to become unsafe in congested traffic.

Multiple ML models generate diverse learned trajectories in a tree structure, cutting heuristic planning cost and speeding vehicle decisions.

A switchable exterior confirmation unit retracts during Level 3+ autonomous driving to cut drag and wind noise without losing needed visibility.

Safety-score evaluation lets an autonomous vehicle trigger discretionary lane changes from traffic, restriction, and neighboring object data.

Target speed control slows automated vehicles in curved sections with oncoming traffic to reduce passenger uneasiness and harsh deceleration.

Road-edge and sign distance rules help ADAS judge whether stop or yield signs apply, reducing unnecessary braking and safety risks.

Vehicle cameras and displays enable legally compliant driver interaction with remote people by capturing, sending, and showing timed images.