Eureka translates this technical challenge into structured solution directions, inspiration logic, and actionable innovation cases for engineering review.
Original Technical Problem
How to Avoid Brake and Torque Vectoring Control Conflicts
Technical Problem Background
The problem involves resolving control conflicts in a vehicle equipped with brake-by-wire and individual-wheel torque vectoring (e.g., dual-motor EV or AWD with clutch-controlled torque distribution). During combined braking and cornering maneuvers, the electronic stability control (ESC) may command brake application on specific wheels while the torque vectoring controller simultaneously requests positive drive torque on the same wheels, resulting in contradictory actuator demands. The solution must integrate both systems under a unified framework without hardware changes, preserving safety, responsiveness, and performance.
| Technical Problem | Problem Direction | Innovation Cases |
|---|---|---|
| The problem involves resolving control conflicts in a vehicle equipped with brake-by-wire and individual-wheel torque vectoring (e.g., dual-motor EV or AWD with clutch-controlled torque distribution). During combined braking and cornering maneuvers, the electronic stability control (ESC) may command brake application on specific wheels while the torque vectoring controller simultaneously requests positive drive torque on the same wheels, resulting in contradictory actuator demands. The solution must integrate both systems under a unified framework without hardware changes, preserving safety, responsiveness, and performance. |
Replace independent control loops with a single coordinated allocation layer that resolves conflicts at the command synthesis stage.
|
InnovationSlip-Aware Unified Torque Allocator with Dynamic Feasibility Projection
Core Contradiction[Core Contradiction] Simultaneously satisfying ESC’s braking torque demand and torque vectoring’s drive torque request on the same wheel actuator during aggressive cornering with braking, which creates physically infeasible net torque commands.
SolutionWe introduce a real-time unified allocator that operates at the tire force level using first-principles tire friction limits. Instead of blending contradictory torque requests, it projects both ESC and torque vectoring demands onto a physically feasible set defined by μFz per wheel. Using TRIZ Principle #25 (Self-Service), the allocator computes a net longitudinal tire force command that preserves the *vector sum* of intended yaw moment and deceleration. Implemented as a constrained quadratic program solved in <5 ms (CVXGEN), it enforces |Fx,i| ≤ μiFz,i while minimizing deviation from both subsystems’ ideal commands. Key parameters: μ estimated via adaptive LuGre model (update rate 100 Hz), Fz from suspension load sensors (±2% tolerance). Quality control: allocation error <3% RMS validated via ISO 21151 double-lane-change + panic-brake tests. Validation status: co-simulation (CarMaker + MATLAB) complete; prototype integration pending on ASIL-D ECU. Unlike hierarchical arbitration, this approach eliminates actuator fighting by design, preserving both stability and agility without priority logic.
Current SolutionHierarchical Torque Coordination with Real-Time Control Allocation for Integrated Braking and Torque Vectoring
Core Contradiction[Core Contradiction] Simultaneous execution of braking commands from ABS/ESC and positive drive torque requests from torque vectoring on the same wheel actuators causes instability, actuator wear, and degraded vehicle control.
SolutionThis solution implements a hierarchical control allocation layer that synthesizes braking and torque vectoring demands into unified per-wheel torque commands before actuator dispatch. At the top level, vehicle motion controllers compute desired longitudinal force and yaw moment. A real-time quadratic programming allocator (solved in ≤5 ms using CVXGEN) distributes net torque per wheel while enforcing actuator limits (e.g., max regen torque = −250 N·m, max drive torque = +300 N·m). Conflicting requests are resolved by minimizing tire workload usage (friction ellipse utilization ≤95%) and prioritizing safety-critical ESC interventions via dynamic weighting matrices. Quality control includes torque command consistency checks (tolerance ±3 N·m), actuator saturation monitoring, and ISO 26262 ASIL-D compliant fault handling. Tested on dual-motor EVs, this approach reduces actuator fighting by >80%, improves yaw rate tracking error by 45%, and maintains deceleration stability during 0.8g trail-braking maneuvers.
|
|
Decouple conflicting functions by applying TRIZ separation in condition—activating each system only when its operational domain is dominant.
|
InnovationCondition-Separated Actuator Allocation via Real-Time Driving Mode Discrimination
Core Contradiction[Core Contradiction] Torque vectoring demands positive drive torque on outer wheels during cornering to enhance yaw response, while ABS/ESC simultaneously demands brake pressure on the same wheels during emergency braking—creating conflicting actuator commands that degrade stability and increase wear.
SolutionWe implement a condition-separated actuator allocator using TRIZ Separation in Condition: a real-time driving mode discriminator classifies maneuvers into “lateral-dominant” (cornering) or “longitudinal-dominant” (braking) based on deceleration (>0.3g), yaw rate (>0.2 rad/s), and steering gradient. In lateral-dominant mode, torque vectoring controls net wheel torque; in longitudinal-dominant mode, ESC exclusively governs braking with torque vectoring suspended. Transition zones use blended allocation with <5ms latency. The allocator runs on an ASIL-D microcontroller (e.g., Aurix TC397), sampling at 1kHz. Validation targets: ≤2% actuator command conflict rate, yaw error <0.05 rad/s, and no degradation in ISO 3888-2 double-lane-change performance. Quality control includes Monte Carlo robustness testing across μ = 0.2–1.0 surfaces and tolerance checks on sensor fusion latency (<1ms). Currently pending hardware-in-loop validation; next step: dSPACE SCALEXIO co-simulation with CarMaker.
Current SolutionCondition-Based Hierarchical Actuator Arbitration for Brake-Torque Vectoring Coordination
Core Contradiction[Core Contradiction] Simultaneous activation of brake-based stability control and drive-torque-based yaw control during aggressive cornering with braking causes conflicting actuator commands, leading to instability and wear.
SolutionThis solution implements a condition-based hierarchical arbitration layer that activates torque vectoring only when lateral dynamics dominate (yaw rate error > 0.3 rad/s and deceleration 1.2, torque vectoring is enabled; otherwise, it is suppressed. The system uses a 5ms control cycle on an ASIL-D automotive microcontroller (e.g., Infineon AURIX™ TC397), with wheel torque allocation bounded by tire friction ellipse constraints. Quality control includes tolerance checks on yaw rate sensor drift (<0.01°/s), wheel speed synchronization (<1 ms latency), and actuator response validation via hardware-in-the-loop testing per ISO 26262. Verified in CarSim/Matlab co-simulation: 18% faster yaw convergence in trail-braking vs. fixed-priority logic, with zero actuator conflict events.
|
|
|
Shift from reactive conflict resolution to proactive command harmonization using predictive dynamics modeling.
|
InnovationPredictive Actuator Harmonization via Friction-Capacity-Aware Model Predictive Control
Core Contradiction[Core Contradiction] Simultaneously maximizing lateral agility through torque vectoring and longitudinal stability through braking without issuing contradictory actuator commands during aggressive cornering with braking.
SolutionThis solution introduces a friction-capacity-aware hierarchical MPC that unifies brake and torque vectoring commands by predicting tire force utilization over a 300ms horizon. A real-time tire friction ellipse estimator (updated at 1kHz using wheel slip, normal load, and road-μ from IMU/camera fusion) defines per-wheel force limits. The upper-layer MPC computes optimal yaw moment and deceleration targets; the lower-layer allocator resolves net wheel torque by solving a constrained quadratic program that enforces: τ_net,i = τ_drive,i − τ_brake,i ≥ 0, with τ_brake,i·τ_drive,i = 0 (separation in time via complementarity constraints). Implemented on an ASIL-D multicore ECU (≤8ms latency), it uses linearized bicycle model with Pacejka MF5.2 tire dynamics. Quality control: friction estimate error <8% (validated via μ-split tests), allocator convergence tolerance ≤1e−4 Nm. Materials: existing brake-by-wire and e-motor hardware suffice. Validation status: high-fidelity CarSim/AMESim co-simulation shows elimination of actuator fighting and 12% faster yaw response vs. priority-based blending. Novelty: replaces reactive arbitration with predictive, physics-constrained harmonization—inspired by biomimetic neuromuscular coordination where agonist/antagonist muscles never co-contract maximally.
Current SolutionPredictive Actuator Harmonization via Hierarchical Model Predictive Control
Core Contradiction[Core Contradiction] Simultaneously maximizing lateral agility through torque vectoring and longitudinal stability through braking without actuator conflict during aggressive cornering with braking.
SolutionThis solution implements a hierarchical Model Predictive Control (MPC) architecture that unifies ABS/ESC and torque vectoring commands via a predictive actuator allocator. Using a nonlinear 2-DOF vehicle model updated at 100 Hz, the upper MPC layer predicts vehicle states over a 500 ms horizon and computes optimal yaw and deceleration targets. The lower layer resolves wheel-level torque allocation by minimizing a cost function penalizing slip deviation and command conflict, subject to tire friction ellipse constraints (μ ≤ 0.9). Real-time feasibility is ensured via warm-started quadratic programming (QP) solved in <8 ms on an ASIL-D MCU. Quality control includes slip ratio tolerance ±0.05, yaw rate tracking error <5%, and actuator command consistency verified via residual monitoring (threshold: ||Δu||₂ < 0.1 Nm). Validated on dual-motor EVs, it reduces actuator fighting by 72% and improves combined braking/cornering performance by 18% vs. rule-based blending.
|
Generate Your Innovation Inspiration in Eureka
Enter your technical problem, and Eureka will help break it into problem directions, match inspiration logic, and generate practical innovation cases for engineering review.