Robot Motion Control Using Virtual Forces for Obstacle Avoidance

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Collaborative robots face challenges in safely performing tasks in shared work environments without colliding with objects or operators, as existing technologies lack effective methods for real-time obstacle detection and trajectory adjustment.

Innovation Solution

A robot controlling method and motion computing device that utilize a depth camera to obtain depth images, generate attractive, repulsive, and virtual force parameters, and output control signals to drive the robot, ensuring safe operation by avoiding collisions while maintaining the original task trajectory.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If a collaborative robot operates in a shared work environment with operators and objects, then the robot can perform tasks in close proximity to humans and objects, but the risk of collision with objects or operators increases

Engineering Contradiction:

Improveability to operate in shared environmentVSAvoidcollision risk

Core Design Contradiction:

Adaptability or versatilityVSObject-affected harmful factors

Solution Approach 1:

The system performs preliminary detection of objects and operators in the workspace using depth cameras before the robot executes its task trajectory. By identifying potential obstacles in advance and calculating repulsive forces based on their positions, the system proactively prevents collisions rather than reacting after contact occurs. This allows the robot to operate safely in shared environments by preparing avoidance maneuvers before conflicts arise.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent introduces virtual force fields (attractive forces toward target trajectory and repulsive forces away from obstacles) as intermediary mathematical constructs that mediate between the robot's task objectives and safety constraints. These virtual forces act as a computational layer that translates physical presence of objects and operators into guidance signals, enabling the robot to navigate shared spaces safely while maintaining task performance.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Reliability

If the robot adjusts its trajectory in real-time to avoid obstacles, then collision avoidance capability improves, but the complexity of the control system increases

Engineering Contradiction:

Improvecollision avoidanceVSAvoidcontrol system complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent replaces complex mechanical collision avoidance systems with a computational field-based approach. Instead of using multiple sensors, mechanical barriers, or complex motion planning algorithms, the system uses virtual force fields calculated from simple depth image data. The attractive force toward the target trajectory and repulsive force from obstacles are computed using straightforward mathematical formulations, significantly reducing control system complexity while maintaining reliable collision avoidance.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system dynamically adjusts the robot's motion parameters by modifying the virtual force parameters (attractive force magnitude and repulsive force magnitude) based on real-time object positions. When obstacles are detected closer to the robot, the repulsive force parameter increases automatically, causing the robot to deviate from its original trajectory. This parameter-based control simplifies the complexity compared to geometric path planning while achieving reliable obstacle avoidance.

Inventive Principle:

Principle #35Parameter changes

3Manufacturing precision

If the robot maintains high trajectory accuracy while avoiding obstacles, then task performance improves, but the response time for obstacle detection and avoidance may be reduced

Engineering Contradiction:

Improvetrajectory accuracyVSAvoidresponse time

Core Design Contradiction:

Manufacturing precisionVSLoss of time

Solution Approach 1:

The system continuously captures depth images and recalculates virtual force parameters throughout the robot's motion, rather than performing discrete obstacle checks. This continuous operation ensures that the robot maintains its target trajectory when no obstacles are present while being constantly prepared to deviate when obstacles appear. The uninterrupted computation of attractive and repulsive forces enables both high trajectory accuracy and rapid response to emerging obstacles without sacrificing one for the other.

Inventive Principle:

Principle #20Continuity of useful action

Data Source

PatentUS11148293B2Motion computing device, robot system and robot controlling method

Publication Date: 2021.10.19 NAT CHIAO TUNG UNIV

AI summary

A robot controlling method includes: obtaining a depth image by a depth camera; receiving the depth image and obtaining an object parameter according to the depth image by a processing circuit; generating an attractive force parameter according to a target trajectory by the processing circuit; generating a repulsive force parameter according to a first vector between the object parameter and a robot by the processing circuit; generating a virtual force parameter according to a second vector between the object parameter and the robot by the processing circuit; and outputting a control signal to the robot to drive the robot according to the attractive force parameter, the repulsive force parameter and the virtual force parameter by the processing circuit.