Autonomous Trailer Hitching Using Camera and IMU Pose Estimation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current systems for autonomously maneuvering a tow vehicle to hitch with a trailer lack efficient methods for determining the trailer's position and orientation relative to the tow vehicle, limiting their ability to accurately align and connect.

Innovation Solution

The method involves using data processing hardware to receive images from cameras and inertial measurement unit data, determining pixel-wise intensity differences, and calculating camera and trailer poses in a world coordinate system to plan and execute a path for autonomous maneuvering, allowing the tow vehicle to align with the trailer hitch.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Extent of automation

If traditional manual hitching methods are used, then the hitching process requires human intervention and judgment, but the accuracy and efficiency of alignment are limited by operator skill and time consumption

Engineering Contradiction:
Improveautonomous hitchingVSAvoidsensor fusion system
Core Design Contradiction:
Extent of automationVSDevice complexity

Solution Approach 1:

The patent combines multiple sensor types (cameras, LIDAR, radar, ultrasonic sensors, IMU) into a unified sensor fusion system that processes data from all sources simultaneously to determine trailer position and orientation, enabling autonomous hitching while managing system complexity through integrated processing

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent introduces an intermediary processing system that fuses data from multiple sensors and translates it into actionable positioning information, acting as a mediator between raw sensor data and the autonomous control system that maneuvers the tow vehicle

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If simple image processing is used, then the processing speed is fast, but the precision in determining trailer position and orientation is insufficient

Engineering Contradiction:
Improvetrailer pose determinationVSAvoidimage processing algorithm
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces traditional mechanical or simple optical alignment methods with advanced image processing algorithms that analyze pixel-wise intensity differences and feature point correlations to precisely determine trailer pose, substituting computational complexity for mechanical simplicity

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent transitions from two-dimensional image analysis to three-dimensional pose determination by incorporating pixel-wise intensity difference analysis across multiple images and correlating feature points in space and time, adding a temporal and spatial dimension to the measurement process

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

3Reliability

If multiple sensors are integrated for accurate trailer detection, then the measurement precision improves, but the system complexity and computational load increase

Engineering Contradiction:
Improvetrailer detection accuracyVSAvoidsensor integration system
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent creates a multi-functional sensor fusion system where a single processing unit handles data from cameras, LIDAR, radar, ultrasonic sensors, and IMU, enabling the system to perform multiple detection functions (position, orientation, distance, motion) through one integrated architecture rather than separate dedicated systems

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentEP3787911B1Trailer detection and autonomous hitching
Publication Date: 2024.10.16 CONTINENTAL AUTONOMOUS MOBILITY US LLC
  • EP3787911B1 patent drawingFigure 1A
  • EP3787911B1 patent drawingFigure 1B
  • EP3787911B1 patent drawingFigure 1C

AI summary

A method for autonomously maneuvering a tow vehicle (100). The method includes receiving images (143) from one or more cameras (142) positioned on a back portion of the tow vehicle (100) and receiving sensor data from an inertial measurement unit (144) supported by the tow vehicle (100). The method also includes determining a pixel-wise intensity difference between a current received image and a previous received image. The method includes determining a camera pose and a trailer pose with respect to a world coordinate system. The camera pose and the trailer pose are based on the images (143), the sensor data (145), and the pixel-wise intensity difference. The method includes determining a tow vehicle path based on the camera pose and the trailer pose. The method also includes instructing a drive system (110) supported by the tow vehicle (100) to autonomously maneuver along the tow vehicle path in a reverse direction causing the tow vehicle (100) to hitch with the trailer (200).