Disinfection robots

Patent metadata

Field	Value
Publication number	US-12564951-B2
Application number	US-202217969850-A
Country	US
Kind code	B2
Filing date	Oct 20, 2022
Priority date	Oct 21, 2021
Publication date	Mar 3, 2026
Grant date	Mar 3, 2026

Abstract

A UV based surface disinfection system that consists of the UV light source, a robot arm, and an omni directional mobile base. The mobile robot can be programmed autonomously and be able to bring the UV light source to the centimeters away from surfaces to achieve effective and efficient surface disinfection. The mobile robot can navigate autonomously in a complicated environment to perform disinfection operation in a large area.

First claim

The invention claimed is:

1 . A disinfection robot comprising: a mobile base with omni-directional wheels; a universally positionable manipulator with an end effector (E.E.); a UV lamp module attached to the E.E. of the manipulator that includes a UV disinfection lamp and a ranging module using sensors to detect the distance of the lamp from a surface to be disinfected; and an electronic control system for planning a path for the base to move about a space and the UV disinfection lamp to be moved over surfaces in the space based on information from the sensors and for controlling the directional wheels of the base and the movement of the manipulator so as to cause the UV disinfection lamp to follow the path and disinfect the space; wherein the electronic control system includes a path processor configured to plan a path by executing instructions according to an object perceptive local planner method comprising the steps of: segmenting X(t) segments from a real-time point cloud; segmenting X re f(s) segments from a global map as the E.E.'s desired path; using the selected segments X(t) with a point cloud sliced Wasserstein based perceptive motion local planner to map a UV desired path compared to an actual UV path; updating the comparison; and updating a speed profile of the E.E. along a geodesics path, whereby an optimal transport path from X(t) to X re f(s) is established.

2 . The disinfection robot according to claim 1 wherein the ranging module includes a multi-line laser scanner and a 360 degrees camera.

3 . The disinfection robot according to claim 2 wherein the ranging module is a LIDAR system.

4 . The disinfection robot according to claim 1 further including a remote processor with a transceiver and a transceiver connected to the electronic control circuit so the remote processor and electronic control circuit can communicate over a transmission link so that a user can control the robot remotely.

5 . The disinfection robot according to claim 4 wherein the remote processor and the electronic control circuit communicate over a WiFi transmission link.

6 . The disinfection robot according to claim 1 wherein the universally positionable manipulator has 3 rotational joints, which are located in three XYZ joint frames, and the mobile base is located in a separate XYZ frame.

7 . The disinfection robot according to claim 1 wherein the electronic control system acquires from the sensors a velocity, acceleration, position and time in task space according to known initial and end points, and generates a plan for a trajectory to control the E.E. so it follows this trajectory.

8 . The disinfection robot according to claim 7 wherein the trajectory is divided into three sections: third-order, fifth-order and third-order trajectories; and after performing the third-order, fifth-order and third-order trajectories in the planner, a pose of each point in task space is transformed into joint space to control the robot's joint through an inverse kinematics and a Jacobian matrix so that commands can be published to control the joints based on the position and velocity, where the velocity and position correspond to a specific time.

9 . The disinfection robot according to claim 1 wherein the electronic control system contains a feedforward part and a feedback part and a controller for the manipulator, wherein the feedforward part is acquired through a Dynamics model to compute torque in the controller and the feedback part is the position and velocity, wherein based on
a position and velocity error, a position loop and a velocity loop are constructed, wherein a feedforward of torque is combined with the position loop and velocity loop to calculate a decoupling output, a controller output is a controller torque output whose value satisfies a requirement of an entire robot system in order to realize a target, including mutual influence between each joint; and wherein the controller torque output is provided to a driver for the manipulator, and the driver generates current to make manipulator motors operate according to the corresponding controller torque output.

10 . The disinfection robot according to claim 1 wherein the comparisons are updated at a rate of 10 Hz and the speed profile is updated at a rate of 50 Hz.

11 . The disinfection robot according to claim 1 further including a time-of-flight (ToF) camera attached to the UV light module at the E.E. and a Light Detection and Ranging (LiDAR) device on the robot base, and wherein the electronic control system includes a path processor configured to plan a path by executing instructions according to the method of planning the path comprises the steps of: receiving a present image, l(t), a present point cloud X(t) and camera_intrinsics.yaml & camera_toflf.launch signals from the robot at a present point cloud segmentation processor; producing at the present point cloud segmentation processor a visual cone in the local frame and X seg (t), which is delivered to a sliced perceptive object tracker; receiving UVdesired.txt, Map.ply and pnum(s) at a.reference point cloud generator, which in turn generates X′ seg (s), which is applied to the sliced perceptive object tracker along with an E.E. State and E.E. constraints, wherein the E.E. constraints come from joint dynamic constraints and environmental information after passing through a dynamics module; causing the sliced perceptive object tracker to engage in trajectory planning with dynamic constraints on a geodesics path and Wasserstein barycentre as intersections from X to X,′ wherein the sliced perceptive object tracker produces as an output a point cloud registration (based on sliced Wasserstein distance); passing the point cloud registration to a point cloud sliced Wasserstein-based perceptive motion local planner, wherein the planner has its output tf: from/UV to/UV desired directed to the reference point cloud generator as an input along with X′ seg (s) and LIDAR_FoV.yami; applying a second output of the planner to a calculation block which engages in e_calculation (known/map to/UV), which provides the input Pnum(s) to the reference point cloud generator; and applying an output p(T) from the sliced perceptive object tracker to a kinematics unit to create q(t), which is directed to the robot, and which in turn provides the present image and present point cloud applied to the present point cloud segmentation unit, whereby global localization is used as an initial condition, then it mainly compares reference point cloud and present point cloud in the end effector frame (E.E.F.) of the mobile manipulator based on Wasserstein distance, and then outputs rigid transformation between them at a low frequency and desired velocity along the optimal path of the E.E. at a high frequency.

12 . The disinfection robot according to claim 1 for conducting a disinfection task with respect to a movable object, further including a 3D camera on the E.E. of the robot, and wherein the electronic control system includes a disinfection motion planning module with a Local Reference Frame (R_(L.R.F)) attached to the movable object, which planning module includes a plan processor configured to execute instructions to carry out the steps of′ obtaining a reference point cloud in advance; collecting a target point cloud with the 3D camera; utilizing a vector space with a point cloud registration algorithm to output a transformation matrix in the vector space based on the reference point cloud and the target point cloud, wherein the transformation matrix is acquired for converting from the reference point cloud to the target point cloud; transferring the original motion planning that is attached to the movable object t

Assignees

• Versitech Ltd

Inventors

• Xi Ning
• Ma Ye
• Wang Siyu
• Wang Qingyang
• Ye Jiajie
• Kwok Chun
• Bi Sheng

Classifications

• B25J13/006

  by means of a wireless system for controlling one or several manipulators · CPC title

• A61L2202/14

  Means for controlling sterilisation processes, data processing, presentation and storage means, e.g. sensors, controllers, programs · CPC title

• A61L2/10

  Ultraviolet [UV] radiation · CPC title

• A61L2/24

  Apparatus using programmed or automatic operation · CPC title

• B25J19/02

  Sensing devices · CPC title