Method and Apparatus for Generating Autonomous Driving Paths

This technology is an algorithm that generates waypoint-based work paths and auxiliary paths from GNSS-based manual driving data, then merges and converts them into network data to calculate the shortest autonomous driving path.

Building autonomous driving paths for agricultural machinery is costly, and there is a lack of methods that allow for universal operation across various zones without the need for high-precision maps.

This technology is an apparatus and method for generating paths that creates waypoints based on location data acquired from positioning sensors, defines each segment as start, end, straight, or turn, and merges multiple work and auxiliary paths to generate the shortest network path. It can be applied to logistics transport, service robots, and autonomous driving platforms, reducing labor costs and improving the efficiency of agricultural operations through autonomous driving.

‍

‍

Key Features:

• Generating a waypoint-based work path using first location data of an object collected from a positioning sensor while the object is being manually driven along the path to be autonomously navigated.
• Generating a waypoint-based auxiliary path, which is a path other than the work path, using second location data of the object collected from a positioning sensor while the object is being manually driven.
• Generating a portion of the first location data as waypoints by using the angle between straight lines to determine whether a segment is straight.
• An autonomous driving path generation method comprising the step of applying a reference angle to define turning segments for waypoints, thereby designating the type of each waypoint as a start point, end point, straight segment, or turning segment.

‍

This invention was developed with support from the Ministry of Science and ICT for the development of a universal unmanned agricultural machinery platform and system based on high-precision positioning technology for open-field smart farms.

Guidewire Microrobot

This technology features a guidewire microrobot structure that is driven and steered by an external magnetic field. It is sealed by welding or soldering both ends of a hollow tube to a metal body, a core wire, and an external coil, with magnetic materials and flexible substances hermetically enclosed inside.

Conventional magnetic guidewires pose clinical safety risks, such as the potential leakage of internal magnetic particles into the body if the flexible polymer is damaged, or the wire breaking or detaching during procedures due to structural weaknesses.

This technology inserts a metal body into one end of the tube and a core wire into the other, wraps the outer surface with a coil, and then uses welding or soldering to physically seal both ends. This prevents the leakage of magnetic materials and flexible substances while enhancing structural integrity. Applicable to industrial robots and automated systems, the interference fit between the metal components and the coil seals one end of the tube, thereby improving clinical safety and preventing detachment.

‍

‍

Key Features:

• The length of the section within the tube containing the flexible substance is longer than the section containing the magnetic material, with the core wire inserted into the other end of the tube.
• A guidewire microrobot characterized in that the portion of the core wire inserted into the tube is positioned within the flexible substance and is spaced apart from the magnetic material.
• The metal body and coil are fixedly coupled to seal one end of the tube, with the flexible substance housed inside the tube.
• A magnetic material disposed between the metal body and the flexible substance.

‍

This invention was developed with support from the Ministry of Health and Welfare for the development of a microrobotic guidewire system for peripheral vascular intervention.

Flexible Hinge Component

This technology features a spherical robot equipped with multiple flexible hinge components, each featuring an X-shaped cross-beam or spring structure on its exterior. These components absorb impact from all directions upon impact, while internal links and actuators allow the robot to switch between throwing mode (spherical) and movement mode (separated/rotating).

Conventional throwable robots are complex to manufacture, often combining multiple leaf springs and rubber supports. While they may absorb lateral impacts, they often fail to mitigate vertical shocks, leaving wheels and internal electronic components vulnerable to damage.

This technology utilizes flexible hinge components made of elastic X-shaped cross-beams or springs, arranged in a grid pattern on the spherical housing to ensure omnidirectional shock absorption. A control unit processes real-time speed and orientation data from sensors to adjust internal counterweights, ensuring the shock-absorbing sections are positioned to mitigate anticipated impact points. Applicable to industrial robots and automation systems, this design provides a throwable, small-scale spherical robot with enhanced shock absorption and a specialized housing for movement and rotation, improving upon the monitoring and cost-efficiency of existing robotic solutions.

‍

‍

Key Features:

• A plate-shaped buffer surface that receives impact and a plate-shaped support surface spaced apart from it
• A hinge section connected to the buffer and support surfaces to dampen the force transmitted from the buffer surface
• The hinge section includes a spring with one end coupled to the buffer surface and the other to the support surface
• A structure where a shock-absorbing unit, formed by linearly coupled flexible hinge components, is attached to the exterior of the robot housing with a gap

Robot Arm Control Device and Method

This technology is a grasping control method for robot arms that uses an optical tracker and a world marker placed within the workspace as a reference point. By calculating coordinate system transformation matrices between the robot tool tip, the world marker, and the target marker in multiple stages, this geometric control method determines the precise target position of the tool tip, even if the tracker's position changes or the robot and target are not within the same field of view.

Conventional eye-to-hand calibration requires a fixed tracker position, which limits the flexibility of the work environment. Furthermore, technologies that require the robot marker and target marker to be recognized simultaneously within a single field of view have significant limitations regarding the operating range.

This technology derives a third transformation relationship by converting the tool tip coordinates of the robot arm and the position coordinates of the target marker based on a world marker. By using a world marker with a polyhedral or curved structure that is easily recognizable from multiple angles, it increases the installation flexibility for the optical tracker and controls robot motion by establishing relational expressions between the robot base, arm end, tool tip, and robot marker. It can be applied to logistics picking, service robots, and manufacturing automation, improving robot arm control in diverse environments and allowing for clear marker recognition from various directions.

‍

‍

Key Features:

• A method where at least a portion of each step is performed by a processor, comprising the step of determining a first transformation relationship, which is the coordinate system transformation relationship between the world marker and the robot arm's tool tip
• The step of determining a fourth transformation relationship, which is the coordinate system transformation relationship between the robot marker located on the robot arm's end effector and the tool tip
• The step of determining a second transformation relationship, which is the coordinate system transformation relationship between the world marker and the target marker located on the target
• The step of determining a fifth transformation relationship, which is the coordinate system transformation relationship between the robot marker and the arm end portion

‍

This invention was developed with support from the Ministry of Trade, Industry and Energy for the development of intelligent controller technology applicable to various commercial articulated robots and specialized for bin-picking and loading/unloading tasks.

Floor Finishing Material Installation System

This technology features a mobile platform equipped with a pivotable guide mechanism that travels between floor leveling materials. By pressing the guide rollers against the materials when moving off the previously installed finishing area, the system maintains platform stability and orientation for autonomous navigation.

Traditional manual floor installation suffers from inconsistent quality and safety risks for workers. Furthermore, existing dedicated robots are limited to specific architectural environments, resulting in poor versatility, increased costs, and reduced productivity due to the need for separate robots for different tasks.

This technology incorporates a material feed opening and an internal lowering device at the rear of the mobile platform, along with a pivotable guide mechanism at the front that inserts into and rolls along the gaps between floor leveling materials. This design assists with autonomous movement and orientation on top of previously installed materials, automating the sequential installation process. Applicable to rehabilitation training, gait assistance, and medical/welfare services, it enhances construction efficiency, productivity, safety, and cost-effectiveness in the field of finishing material installation.

‍

‍

Key Features:

• A mobile guide unit that inserts into and rolls along the gap between two adjacent floor leveling materials as the drive shaft pivots downward.
• Multiple obstacle detection sensors installed along the sides of the mobile platform to identify obstructions.
• Bumper sensors installed around the lower perimeter of the mobile platform to detect collisions.
• A mobile platform featuring an elongated rear opening for feeding floor finishing materials, which are then lowered onto the pre-installed floor leveling materials inside the body.

‍

This invention was developed with support from the Ministry of Science and ICT for the development of intelligent painting and masking collaborative robots.

Biological Material Transport Device

This technology features a capsule-type robot structure consisting of a body with multiple internal chambers and openings, and a magnetic drive unit that moves forward, backward, and rotates via an external magnetic field to press the transport mechanism of a selected chamber. Through the mechanical interaction between guide grooves and guide pins, it independently controls the protrusion of multiple transport mechanisms to perform the collection or delivery of biological materials.

Conventional in-vivo transport capsules rely on peristalsis, making active movement impossible, and lack the functionality to operate selectively while preventing cross-contamination during multiple sample collections or multi-material delivery.

This technology employs a selective drive mechanism where transport mechanisms are housed in multiple chambers arranged radially within the body, and a rod on the magnetic drive unit—movable and rotatable along a shaft via an external magnetic field—presses and extends a specific transport mechanism. Applicable to surgical robots, interventional systems, and medical automation, it improves the efficiency of collecting samples from specific locations within the digestive tract and enhances independent multi-picking and delivery capabilities.

‍

‍

Key Features:

• Multiple transport mechanisms housed in individual chambers for storing biological materials, capable of protruding outside the body through openings.
• A magnetic drive unit that moves within the body via an external magnetic field to selectively press one of the multiple transport mechanisms.
• A body with multiple openings formed on its rear surface and multiple chambers corresponding to each opening formed inside.
• A biological material transport device comprising a ring-shaped first magnetic body with a through-hole and a rod protruding from the first magnetic body toward the chamber housing the transport mechanism.

‍

This invention was developed with support from the Ministry of Health and Welfare for the development of therapeutic modules for micro-medical robots.

Avoidance Path Generation Method and Control Device Based on Multi-Sensor Fusion Using Control Infrastructure

This technology synchronizes and fuses data from fixed sensors installed in control infrastructure with onboard sensor data from Autonomous Mobile Robots (AMRs) based on a spatio-temporal grid. By doing so, it detects hazards in the robot's blind spots and generates avoidance paths.

AMRs often face collision risks because their onboard sensors have limitations, making it difficult to detect objects located behind obstacles or within blind spots in advance.

This technology establishes a path-generation control system that receives data from control infrastructure sensors (cameras, LiDAR, etc.) and the robot's own sensors, synchronizes them with the robot's location, and fuses them using a grid fusion method to avoid paths containing hazards. Applicable to logistics transport, service robots, and autonomous driving platforms, it prevents collisions with obstacles in blind spots and enhances the ability of mobile objects to actively manage objects in their path, thereby improving safety and efficiency across various fields.

‍

‍

Key Features:

• The control device receives second sensing data regarding the surrounding area from a mobile object moving along a set path.
• The control device uses the fused data to determine whether there are any hazards in the area where the mobile object is intended to move.
• The control device receives first sensing data regarding a region of interest from sensors within the control infrastructure.
• If a specific object is present in the region of interest that includes the mobile object's path—based on at least one of the first or second sensing data—the control device fuses the first and second sensing data to generate fused data.

‍

This invention was developed with support from the Ministry of Science and ICT for the development of autonomous avoidance driving technology for the pre-recognition and protection of traffic objects through infrastructure linkage.

Device for Calculating Optimal Surveillance Sensor Placement

This technology is a simulation-based system that calculates the optimal installation locations for surveillance sensors based on movement path scenarios of objects and mobile units within an indoor monitoring area. By simulating the total time during which the detection ranges of surveillance sensors and mobile unit sensors overlap, it identifies the sensor placement configuration that maximizes object detection time.

The risk of collision with objects due to blind spots in autonomous robot detection ranges, and the inefficiencies of existing surveillance sensor installation methods (increased costs due to over-installation or blind spots caused by under-installation).

This technology is a device and method that performs simulations by inputting time-based movement path scenarios for objects and mobile units along with monitoring area maps. It calculates the 'object detection time' for various combinations of potential surveillance sensor locations and identifies the optimal installation index that maximizes this time. Applicable to logistics transport, service robots, and autonomous driving platforms, it improves object detection efficiency, reduces the number of required control sensors, and enhances the overall safety of autonomous robots within the monitored area.

‍

‍

Key Features:

• A location calculation unit that determines the optimal installation information for surveillance sensors based on the calculated object detection time.
• An information receiving unit that receives the number of surveillance sensors, potential installation location data, and map information of the monitoring area.
• A time calculation unit that computes the object detection time, which is the total duration during which surveillance sensors or mobile units detect an object.
• A device for calculating optimal surveillance sensor placement, characterized by the above features.

‍

This invention was developed with support from the Ministry of Science and ICT for the development of an autonomous mobile robot blind-spot avoidance path optimization algorithm based on multi-sensor fusion for efficient manufacturing process automation.

Robot and Method for Estimating Robot Position

This technology is a complex sensing control system that precisely estimates the positions of a robot and surrounding objects by arranging multiple sensor units at equal angles around the robot's exterior and integrating vision, depth, and marker tracking information. It updates real-time position data by converting individual sensor reference coordinates into a single coordinate system centered on the robot.

Existing LiDAR and radar sensors are limited in their ability to perform precise autonomous driving and tasks due to narrow fields of view, low resolution, and limited accuracy. Furthermore, it has been difficult to reliably estimate posture when the robot's position fluctuates.

This technology utilizes a housing containing vision cameras, depth sensors, and marker trackers, arranged around the robot's body to acquire multi-angle environmental information. A processor maps the reference coordinates of each sensor to the robot's origin coordinates and updates map and position data complementarily based on marker recognition to control autonomous navigation. Applicable to robot gripping, precision measurement, and automated facilities, this technology improves the accuracy of position estimation for both the robot and objects in the workspace, thereby enhancing overall operational performance.

‍

‍

Key Features:

• Sensor units disposed on the main body, including a vision camera for capturing images of objects, a depth sensor for acquiring depth information, and a marker tracker for tracking the position of markers attached to objects.
• A processor that receives image, depth, and marker information from the sensor units to derive object position information and uses that information to estimate the robot's position.
• A robot comprising a main body unit that forms the exterior of the robot and includes a drive unit.
• A robot that maps depth information and marker information, each mapped to a plurality of reference coordinates, onto robot coordinates.

Grip Module

This technology features a mechanism that maintains contact with objects and secures gripping force by filling an elastic body with a variable-viscosity fluid (magnetorheological fluid) and controlling the attraction and repulsion between permanent magnets to adjust the fluid's viscosity.

Conventional grippers require custom fabrication to handle objects of various shapes, which reduces operational efficiency and limits the ability to achieve a secure fit based on the object's geometry.

This technology consists of a module that conforms a body containing magnetorheological (MR) fluid to an object, then uses a motor to control the relative positions of first and second permanent magnets. This generates a magnetic field through repulsive force, increasing the fluid's viscosity to lock the body's shape in place. Applicable to logistics picking, service robots, and manufacturing automation, it enhances operational stability and performance by maintaining grip strength and improving the viscosity response of the fluid section.

‍

‍

Key Features:

• A grip module provided on the robot hand of a gripper that picks up and moves objects, comprising a body portion provided on the gripping surface of the robot hand with an internal space.
• A fluid portion provided inside the body portion capable of changing viscosity.
• A motor that drives at least one of the connected first and second permanent magnets to generate attraction or repulsion relative to at least one other of the first and second permanent magnets.
• A second permanent magnet facing the first permanent magnet, with N and S poles each provided in at least one divided pair.

‍

This invention was developed with support from the Ministry of Trade, Industry and Energy for the development of a gripper system for high-mix production processes capable of securely gripping unspecified objects of various shapes, weights, and strengths.

Biomimetic Modular Artificial Muscle

This technology is an artificial muscle control system that applies a catalytic coating to shape-memory alloy (SMA) wires and drives them by generating heat through the catalytic combustion of chemical fuel instead of electrical heating. Fuel supply is controlled via an electromagnetic valve module, and mechanical movement is performed by inducing shape changes in the cantilever-type wire based on signals from a microcontroller.

Conventional SMA-based artificial muscles use Joule heating, which leads to high power consumption and stability issues related to electrical current usage. Furthermore, they are limited in their ability to perform diverse movements due to the short stroke of linear actuation and the lack of integrated fuel storage and precision control systems.

This technology integrates a shape-memory alloy wire coated with platinum black and carbon nanotubes, methanol fuel for chemical heating, and an electromagnetic valve module into a single unit. Upon receiving a pin signal from the control unit, the valve opens to trigger a chemical combustion reaction, which straightens the curved wire to actuate terminal devices such as artificial fingers. Applicable to industrial robots and automation systems, it utilizes chemical power and signal control within an integrated module, improving the stability and efficiency of artificial muscle operation through microcontroller-based control.

‍

‍

Key Features:

• Artificial muscle unit where the shape-memory alloy wire passes through both sides of the substrate, with both ends protruding.
• Electromagnetic valve module that opens and closes the valve using electromagnetic force, including a coil and magnet.
• Actuation structure where the wire is bent to a predetermined curvature to move an artificial finger.
• Control system including a receiver that receives pin signals from muscles or nerves via a control unit.

‍

This invention was developed with support from the Ministry of Science and ICT for the localization of smart electronically controlled lower-limb prosthetics and core components for patients with bilateral lower-limb amputations.

Method and Apparatus for Autonomous Collision Avoidance in Unmanned Aerial Vehicles

This technology implements an autonomous collision avoidance algorithm that assesses collision probability based on the relative distance and velocity of objects surrounding the UAV, and generates flight control commands by combining user control inputs with collision avoidance vector values. In particular, when multiple objects are present, if the directions of the vectors are within a preset angle during the sequential vector summation process, an orthogonal avoidance motion vector is added to prevent inefficient behaviors such as stalling.

Conventional methods based on the Repulsive Potential Field (RPF) calculate collision avoidance forces as infinite values, necessitating converter processing. Furthermore, these methods often result in the UAV stalling due to the cancellation of user control inputs by collision avoidance values, and they suffer from an inability to provide 360-degree omnidirectional detection due to sensor Field of View (FOV) limitations.

This technology includes a collision assessment unit that determines the likelihood of a collision based on the relative distance and velocity of objects near the UAV, and a collision avoidance calculation unit that sums a first collision avoidance vector based on UAV velocity and a second collision avoidance vector based on user control input. When summing the avoidance vectors for the i-th object, if the vectors fall within a preset angle, an avoidance motion value in the orthogonal plane is added to ensure collision avoidance. Applicable to unmanned exploration, surveillance, and environmental monitoring, this technology improves the calculation of collision prevention values by accounting for both user control inputs and UAV velocity.

‍

‍

Key Features:

• A collision assessment unit that determines collision probability based on the shortest distance of a virtual line generated from the relative distance and velocity between the detected object and the UAV.
• A collision avoidance calculation unit that computes collision avoidance vector values only for objects determined to have a collision probability.
• Collision avoidance vector values are calculated by summing a vector based on UAV velocity and a vector based on user control input.
• A configuration that sequentially sums the avoidance vectors of each object and determines the directional angle relative to the previous vector sum.

‍

This invention was developed with support from the Ministry of Science and ICT for smart convergence technology for active defense systems in urban UAV safety management.

Camera-Robot Calibration Method for Automated Labeling

This technology is a camera-robot calibration algorithm that estimates the transformation matrix between the vision sensor's image coordinate system and the robot arm's world coordinate system, improving the precision of coordinate transformation through projection error calculation and updating processes.

Conventional manual calibration methods are time-consuming and prone to errors, as they require repetitive manual tasks to acquire hundreds of feature point pairs.

This technology estimates an initial transformation matrix through primary labeling and performs conditional automated secondary labeling and data association based on projection error, iteratively optimizing the transformation matrix with minimal manual intervention. It can be applied to robot gripping, precision measurement, and automated equipment, thereby improving the accuracy and efficiency of automated labeling systems by implementing a calibration method for precise coordinate estimation and transformation.

‍

‍

Key Features:

• A camera-robot calibration method for automated labeling, comprising the step of analyzing a received image to acquire first center coordinates of labels attached to a first number of objects among a plurality of objects, and first world coordinates for the first center coordinates.
• Based on conditions regarding the magnitude of a first projection error, the step of acquiring second-a center coordinates for a second number of objects among the plurality of objects, and acquiring second world coordinates for the second-a center coordinates using the estimated first transformation matrix.
• The step of analyzing an image received from a vision sensor regarding an attached label to estimate second-b center coordinates of the attached label in the image coordinate system.
• The step of automatically moving a robot arm to the estimated second world coordinates to attach a label to the object at that position.

‍

This invention was developed with support from the Ministry of Science and ICT for the development of a deep learning-based collaborative robot automated round bar labeling system with worker proximity detection.

Steerable Microrobot for Guidewires

This technology features a steering unit containing magnetic materials that respond to external magnetic fields, and a position-tracking unit based on quantum dots that absorb and re-emit short-wave infrared light capable of penetrating the body. These are integrated into the tip of a guidewire to enable non-invasive, real-time tracking and steering.

Conventional vascular intervention procedures rely on X-ray imaging to track guidewire positioning, which exposes both patients and medical staff to radiation and carries risks of side effects from contrast agents, such as shock or heart failure.

This technology incorporates a position-tracking unit at the guidewire tip, consisting of a matrix embedded with magnetic materials and micro/nanoparticles (including quantum dots). This allows for precise steering via external magnetic fields and real-time internal positioning via short-wave infrared illumination. Applicable to industrial robotics and automation systems, this technology eliminates the need for radiation, improves position tracking, removes the risks associated with contrast agents, and enables remote control of the guidewire through complex vascular structures.

‍

‍

Key Features:

• Magnetic material located at the guidewire tip for position control and steering via external magnetic fields
• A connector that links the guidewire to the magnetic material and houses it internally
• A position-tracking unit integrated into the connector that absorbs and emits externally applied light
• The position-tracking unit consists of a matrix containing a mixture of quantum dots and magnetic micro/nanoparticles

‍

This invention was developed with support from the Ministry of Health and Welfare for the development of a microrobotic guidewire system for peripheral vascular intervention.

Magnetic Actuation System with Curved Magnetic Cores

This technology is a magnetic actuation control system that utilizes an electromagnet array with curved magnetic cores to minimize the occupied footprint and concentrate magnetic force toward a target location.

Conventional radial electromagnet arrangements occupy significant space, making them difficult to install due to interference issues with medical imaging equipment such as C-arms, and limiting the precision of magnetic field formation.

By forming the tips of the magnetic cores into a curved shape, this technology allows electromagnets to be arranged on a flat plane while still being oriented toward the target, thereby increasing layout flexibility and improving space efficiency. It can be applied to industrial robots and automation systems to enhance output magnetic fields and magnetic force, allowing for increased control and precision in the movement of self-propelled robots.

‍

‍

Key Features:

• An electromagnet array comprising a plurality of electromagnets aligned in a layout space to generate magnetic fields and magnetic forces
• Each electromagnet consists of a magnetic core and a linearly extended electromagnet body with a coil wound around it
• A magnetic core tip extending from the magnetic core to protrude from the electromagnet body
• The magnetic core tip is curved relative to the magnetic core axis and oriented toward the target position

‍

This invention was developed with the support of the Ministry of Science and ICT’s project for the development of magnetic multi-sequential multi-bot-based neural network reconstruction platform technology.