Driver-Centric Predictive Control for Safe Semi-Autonomous Braking

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional vehicle control systems fail to effectively blend driver intent with a planned trajectory, leading to a need for improved safety and driver-first control strategies in semi-autonomous vehicles.

Innovation Solution

A driver-centric model predictive controller that uses a receding horizon control approach with model predictive control (MPC) to blend driver intent and planned trajectory, applying quadratic cost functions and linear constraints to generate a vehicle command output, ensuring safety by adjusting acceleration and velocity within predefined boundaries.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional vehicle control systems are used, then the system structure is simple, but the system fails to effectively blend driver intent with planned trajectory, compromising safety and control quality

Engineering Contradiction:
ImprovesafetyVSAvoidcontrol system complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The control system is segmented into distinct functional modules: a trajectory generator that creates planned trajectories, a driver intent recognizer that processes driver inputs, and a controller that blends these inputs. This modular segmentation allows each component to specialize in specific tasks, improving overall safety while managing complexity through organized functional decomposition.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The controller acts as an intermediary between the trajectory generator and the vehicle actuators, blending driver intent with planned trajectories. This intermediary component reconciles conflicting inputs by computing a composite control signal that satisfies both safety constraints and driver preferences, thereby improving reliability without requiring complete system redesign.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If autonomous intervention mode is used to ensure safety, then collision prevention is improved, but driver control authority is reduced

Engineering Contradiction:
Improvecollision preventionVSAvoiddriver control authority
Core Design Contradiction:
ReliabilityVSEase of operation

Solution Approach 1:

The control system dynamically adjusts the level of autonomous intervention based on real-time conditions. When driver intent aligns with safe trajectory requirements, the system maintains high driver authority. When potential conflicts arise, the system dynamically increases intervention to prevent collisions, thereby balancing safety with driver control authority through adaptive, condition-dependent behavior.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system continuously monitors driver inputs and vehicle state, providing feedback to the controller that adjusts the blending of driver intent and planned trajectory in real-time. This feedback mechanism ensures that autonomous intervention is applied only when necessary for collision prevention, maintaining driver control authority during normal operation while ensuring safety when needed.

Inventive Principle:
Principle #23Feedback

3Ease of operation

If driver-first control strategy is implemented, then driver control authority is maintained, but safety constraints may be compromised

Engineering Contradiction:
Improvedriver control authorityVSAvoidsafety constraints
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The system preemptively applies safety constraints by blending driver intent with pre-computed planned trajectories that inherently satisfy safety requirements. Rather than reacting to unsafe conditions after they occur, the controller proactively combines driver inputs with safety-guaranteed trajectory references, ensuring safety constraints are maintained while preserving driver control authority through the blending mechanism.

Inventive Principle:
Principle #9Preliminary anti-action

Data Source

PatentUS12366863B2Driver-centric model predictive controller
Publication Date: 2025.07.22 TOYOTA JIDOSHA KK
  • US12366863B2 patent drawing
  • US12366863B2 patent drawing
  • US12366863B2 patent drawing

AI summary

A method for controlling a vehicle includes controlling, at a first time period, the vehicle based on a first velocity profile constrained by an acceleration profile that includes a minimum acceleration, a maximum acceleration, and a maximum velocity. The method also includes generating an adjusted acceleration profile by adjusting one or more of the minimum acceleration, the maximum acceleration, or the maximum velocity associated with the acceleration profile based on determining an object is approaching the vehicle. The method further includes controlling, at a second time period, the vehicle based a second velocity profile constrained by the adjusted acceleration profile.