Object avoidance with object detection and classification

Patent metadata

Field	Value
Publication number	US-10133276-B1
Application number	US-201514744552-A
Country	US
Kind code	B1
Filing date	Jun 19, 2015
Priority date	Jun 19, 2015
Publication date	Nov 20, 2018
Grant date	Nov 20, 2018

Abstract

A robot equipped with an object detection system and an object identification system is used to move inventory holders throughout a warehouse or other environment. The robot can detect objects in its path using the object detection system, which can include one or more sensors for this purpose. The robot can then classify the object using the object identification system to determine an appropriate course of action. The robot can classify the object as an inventory item, warehouse equipment, or a person, among other things. The robot can take action based on the object classification. The robot can reroute around inventory items and warehouse equipment. When encountering people or objects that cannot be classified, the robot can stop and await further instructions. In some cases, the robot may wait for a manual reset before continuing along its path.

First claim

What is claimed is:

1. A method to detect and classify an object in a pathway of a robot, the method comprising:
receiving, with a transceiver on a robot, instructions to travel to a location via a pathway;
sending, from a processor on the robot to a drive system on the robot, instructions to travel along the pathway;
detecting, with an object detection system on the robot, an object in the pathway;
sending, from the processor to the drive system, instructions to slow or stop;
scanning the object with a radio frequency identification (RFID) reader on the robot to determine if there is an RFID tag associated with the object;
scanning the object with a three-dimensional (3D) scanner to provide approximate sizing data for the object to the processor;
scanning the object with an imaging device to provide indicia data to the processor;
determining that the object does not have an RFID tag;
determining, based at least in part on the sizing data that the object is trash; and
sending, from the processor to the drive system, instructions to run over the object.

2. The method of claim 1 , wherein the sizing data indicates that the object is below a predetermined size range.

3. The method of claim 1 , wherein the RFID reader provides no RFID data from the object to the processor;
wherein the sizing data for the object is within a predetermined size range; and
wherein the imaging device provides no indicia data for the object.

4. The method of claim 1 ,
wherein a size of the object is above or below a predetermined range; and
wherein, responsive to the size of the object being above or below the predetermined range, the processor classifies the object as a non-human.

5. A method to detect and classify an object in a pathway of a robot, the method comprising:
receiving, with a transceiver on a robot, instructions to travel to a location via a pathway;
sending, from a processor on the robot to a drive system on the robot, instructions to travel along the pathway;
detecting, with an object detection system on the robot, an object in the pathway;
sending, from the processor to the drive system, instructions to slow or stop;
scanning the object with a radio frequency identification (RFID) reader on the robot to determine if there is an RFID tag associated with the object;
scanning the object with a three-dimensional (3D) scanner to provide approximate sizing data for the object to the processor;
scanning the object with an imaging device to provide indicia data to the processor; determining that the object has an RFID tag associated with an inventory item;

determining that the damage caused by the robot running over the inventory item is below a predetermined threshold; and
sending, from the processor to the drive system, instructions to run over the object based on one or more of whether the RFID tag is present, the sizing data, or the indicia data.

6. The method of claim 5 , wherein the imaging device comprises a video camera; and
wherein the indicia data comprises a surface coating on the object with a predetermined reflectivity associated with one of an inventory item or equipment.

7. The method of claim 5 , wherein the sizing data indicates that the object is above or below a predetermined size range.

8. The method of claim 5 , wherein a size of the object is above or below a predetermined range; and
wherein, responsive to the size of the object being above or below the predetermined range, the processor classifies the object as a non-human.

9. A robot configured to move inventory holders about an environment and comprising:
a drive system configured to move the robot throughout the environment;
an object detection system configured to detect an object in a pathway of the robot;
an object identification system configured to receive classification data for the object;
a transceiver configured to send and receive wired and wireless communications; and
a processor in communication with at least the drive system, object detection system, object identification system, and transceiver, the processor including computer-executable instructions to cause the processor to:
receive, with the transceiver, instructions to travel to a location via a pathway;
send, from a processor to the drive system, instructions to travel along the pathway;
detect, with the object detection system, an object in the pathway;
send, from the processor to the drive system, instructions to slow or stop;
scan the object with a radio frequency identification (RFID) reader on the robot to determine if there is an RFID tag associated with the object;
scan the object with a three-dimensional (3D) scanner to provide approximate sizing data for the object to the processor;
scan the object with an imaging device to provide indicia data to the processor;
determine that the object has an RFID tag associated with an inventory item;
determine that the damage caused by the robot running over the inventory item is below a predetermined threshold; and
send, from the processor to the drive system, instructions to run over the object based on one or more of whether the RFID tag is present, the sizing data, or the indicia data.

10. The robot of claim 9 , wherein the sizing data indicates that the object is above or below a predetermined size range.

11. The robot of claim 9 , wherein a size of the object is above or below a predetermined range; and
wherein, responsive to the size of the object being above or below the predetermined range, the object identification system classifies the object as a non-human.

12. The robot of claim 9 , wherein the imaging device comprises a video camera; and
wherein the indicia data comprises a surface coating on the object with a predetermined reflectivity associated with one of an inventory item or equipment.

Assignees

• Amazon Tech Inc

Inventors

• Shydo Jr Robert Michael
• Caldara Stephen A
• Stubbs Andrew

Classifications

• G05D1/65

  Following a desired speed profile · CPC title

• G05D1/43 Primary

  Control of position or course in two dimensions [2D] · CPC title

• Y10S901/01

  Mobile robot · CPC title

• Y10S901/47

  Optical · CPC title

• G05B2219/31005

  Detect obstacles on path of vehicle · CPC title