Camera pixel classification projected onto LiDAR point clouds automates 3D scene building for vehicle testing with less manual setup.
Multiple simplified 3D tire sectors and steady-state transport analysis predict non-axisymmetric tread wear faster with high accuracy.
Extracts key clips from real autonomous driving tests to recreate authentic simulation scenarios with less data and faster scenario construction.
Multi-layer PCB winding synthesis balances stray capacitance, magnetic loss, and fabrication limits in bidirectional CLLC converters.
Specified flux barrier ratios, spacing, and angles cut reluctance rotor design time while improving power density and reducing torque ripple.
A closed-loop exchange between tire and body dynamics improves unmanned vehicle simulation accuracy and reduces road testing cost.
A loop-based compression network with LSTM and model-driven optimization cuts distribution grid reconfiguration time and computing load.
Real vehicle state and control data train a recurrent model that captures manufacturing variation and improves state prediction accuracy.
A configurable component library generates and compares hybrid powertrain architectures to reduce design bias and complexity.
Perturbing real-world AV log parameters creates realistic simulated scenarios, expanding test coverage without new road data collection.
Targets key vibration transmission frame areas and optimizes local sheet thickness to cut automotive panel noise without changing frame shape or weight.
Virtual driving data and radar logic reproduce AEB warning and braking behavior to estimate stopping distance and collision speed.
Simulated safety, comfort, progress, rules, and occlusion costs help a vehicle choose safer intersection actions without unnecessary delays.
A shared obstacle resource pool builds realistic multi-obstacle interactions for faster, more reliable autonomous driving algorithm testing.
By partitioning module positions and comparing candidate strings, this case cuts PV cable use and avoids repeated manual rework.
Inscribed arc chamfering and pole-shoe height optimization cut AC traction motor electromagnetic noise without changing the rotor.
Real-time drone and vehicle sensing improves rugged-terrain routing, vehicle selection, and map accuracy for remote deliveries.
Array images are used to identify module positions and identifiers automatically, replacing manual code scanning for faster layout mapping.
Centralized sensor data from a global environment model is transformed into vehicle-local coordinates for more accurate AGV control and commissioning.
Automatically generated environmental entities expand traffic-scene diversity and improve continuous autonomous vehicle simulation testing.
Response-surface optimization selects pillar spacing and radius to reduce temperature rise and cooling-liquid pressure drop in a vehicle power module radiator.
Configurable simulated vehicles reproduce lane changes, overtaking, and acceleration to test motion planners with realistic traffic and replayable scenarios.
3D anthropometric modeling replaces fixed negative molds to speed custom seat cushion production while improving fit, comfort, and support.
Replaceable digital modules let a heavy-duty vehicle twin model auxiliary power use in real time and predict power sufficiency.
Truncated parameter regions focus autonomous driving simulations on rare-event scenarios, cutting compute time while improving probability estimates.
Estimate ADS crash probabilities by severity level using an extended limit state function and subset simulation in virtual testing.
Logged path and intent signals are used to build interactive road-user agents that preserve realistic behavior in autonomous driving simulations.
3D-modeled charger inserts match ear-canal hearing instruments to keep charger elements aligned for stable, cost-effective charging.
Real-world agent traces are turned into reactive closed-loop simulations, cutting AV safety testing time while preserving behavior realism.
Optimized stator winding geometry cuts skin- and proximity-effect AC losses, enabling lighter high-frequency electric machines.
Define non-ego vehicle actions relative to the ego vehicle at set time points to model collisions, rotation, and trigger-based behavior more accurately.
Lithographically placed partial air gaps keep laminated magnetic cores flatter in inductance across current ranges while delaying saturation and cutting eddy losses.
Captured command and actuator response data build a vehicle-specific latency model that adds delay and dampening to autonomous driving simulation.
Hierarchical device and component identifiers let a host system verify aircraft installation and quickly flag misinstalled communication-capable parts.
Raw material clustering triggers new equipment evaluation models only when properties differ, cutting processing load and time loss.
Bidirectional workpiece-tool modeling with neural prediction and genetic search improves tool life and surface quality under changing machining conditions.
Parametric tool holder modeling cuts manual setup for third-party holders and supports reliable assembly and interference simulation.
Parametric deformation fitting aligns CAD paths to deformed workpiece point clouds, improving robot path accuracy with lower computation.
Real-time pressure and piston-position data feed a machine learning model to verify hole cold expansion quality without costly residual stress measurement.
AI scores component tolerances to recommend part combinations that cut end-product failures while reducing waste from usable parts.
Neural networks fuse low-capacity vehicle sensors into simulated high-capacity perception data, cutting sensor cost while preserving guidance accuracy.
Analytical kinetic shape equations derive rolling geometries that redirect applied force into target ground reaction forces for shoes and prosthetics.
Separating impact and sliding wear models improves exhaust valve seat wear calculation under varying load, angle, and temperature conditions.
Inverse modeling of PV, OP, and estimated MV quantifies valve stick and jump behavior to cut oscillations and improve loop tracking.
A double-walled pipe and bushing gap improve lubricant delivery while limiting leakage in planetary transmissions used in wind turbine drivetrains.
A hardware-in-the-loop setup combines traffic scenario simulation and real-time interfaces to deliver credible, repeatable cloud control platform testing.
Helical crescent blades with arc transitions improve chip flow while reducing chatter, cutting force, and temperature in aerospace milling.
Real-world driving data is mirrored in a virtual trip so operators can review risky habits and improve safety awareness without losing autonomy.
Pre-installation comparison of spatial and equipment layout data flags noncompliant air-conditioner, lighting, and elevator placement before build.
Layer-specific calculation rules automate structural measurement verification, eliminating manual entry errors and fraud in drywall billing workflows.
A multi-language processor executes design rules across different programming languages to manage complex product models.
Predicted emission plumes guide greedy sensor selection to improve detection coverage while lowering deployment costs for point sources.
An open policy agent bridge system processes entitlement queries using ODRL and DRM data to determine access rights.
Arc length parameterization enables automated detection of high curvature regions on curves without human intervention.
Computational approach nests worst-case load identification with failure minimization to overcome tensile-compressive strength asymmetry.
Modular actuated column array generates configurable 3D topologies, replacing static set construction with dynamic physical reconfiguration.
Hierarchical box groupings in a dynamic layout system organize complex aircraft wiring diagrams, reducing time to locate component connections.
Graphical user interface generates three-dimensional objects from two-dimensional profiles and floor plans.