Autonomous cart for manufacturing and warehouse applications

Patent metadata

Field	Value
Publication number	US-11086328-B2
Application number	US-201916277291-A
Country	US
Kind code	B2
Filing date	Feb 15, 2019
Priority date	Aug 23, 2016
Publication date	Aug 10, 2021
Grant date	Aug 10, 2021

Abstract

An autonomous cart moves products and materials in an industrial environment. It is different from conventional carts because it can navigate autonomously indoors or outdoors in dynamic environments where things change frequently. This autonomous cart uses state-of-the-art “dense” visual perception giving it unequalled and continuous awareness of its surroundings. With this it can operate at a cost, speed, level of safety and efficiency that has never been possible before. This robotic cart makes factories and warehouses more efficient and safer. It enables the movement of smaller batches of material more frequently, reduces the need for expensive conveyor systems, and helps eliminate dangerous and polluting fork trucks from indoor environments.

First claim

The invention claimed is:

1. An autonomous cart comprising:
a chassis;
a plurality of wheels, mechanically coupled to the chassis, to enable movement of the chassis along a surface;
a motor, mechanically coupled to at least one wheel in the plurality of wheels, to rotate the at least one wheel and to turn the at least one wheel to slow and/or stop the autonomous cart;
a load cell, mechanically coupled to the chassis, to sense force exerted on the chassis and to generate a signal representing a magnitude and direction of the force; and
a processor, operably coupled to the load cell, to actuate the motor in response to the signal generated by the load cell;
wherein the autonomous cart is configured to navigate through an environment at a low cost by:
determining a first path of motion to navigate through the environment;
determining a total cost of the first path of motion using a costmap associated with the environment;
determining a second path of motion to navigate through the environment;
determining a total cost of the second path of motion using the costmap;
determining that the total cost of the second path of motion is the low cost; and
selecting the second path of motion.

2. A method, comprising:
receiving, by a first autonomous cart, a map of an environment from a server, the map comprising a costmap that represents a cost to the first autonomous cart for moving through a certain portion of the environment;
determining a swept volume for the first autonomous cart, wherein the swept volume represents a sum of footprints that the first autonomous cart will occupy after navigating through the environment;
navigating, by the first autonomous cart, through the environment based on the costmap at a low cost, wherein the cost is determined using the costmap;
detecting, by the first autonomous cart, a change in the environment with respect to the map while navigating through the environment, the change being at least one of an alteration in visual appearance of the environment with respect to the map or a discovery of a new portion of the environment not represented in the map; and
transmitting, by the first autonomous cart, a representation of the change to the server.

3. The method of claim 2 , further comprising:
receiving, by the first autonomous cart, an updated map of the environment from the server, the updated map showing the change detected by the first autonomous cart.

4. The method of claim 2 , wherein the costmap further represents at least one of a passable area or an impassable area for the first autonomous cart in the environment.

5. The method of claim 4 , wherein navigating through the environment includes:
generating a two-dimensional (2D) traversability map based on the costmap, the 2D traversability map including traversability information for the first autonomous cart.

6. The method of claim 5 , further comprising:
identifying a plurality of destinations for the first autonomous cart based on the 2D traversability map;
evaluating a set of maneuvers for the first autonomous cart to arrive at each of the plurality of destinations;
calculating a feasible path based on the evaluation of the set of maneuvers;
generating a planned path based at least in part on the feasible path; and
obtaining a smoothed trajectory for the first autonomous cart based on the planned path.

7. The method of claim 6 , further comprising:
determining whether the smoothed trajectory violates a constraint in the traversability map;
in response to determining that the smoothed trajectory violates the constraint, rejecting the smoothed trajectory and subsequent trajectories within the planned path; and
in response to determining that the smoothed trajectory does not violate the constraint, executing the smoothed trajectory and subsequent trajectories on the planned path.

8. The method of claim 2 , further comprising:
determining a first path of motion to navigate through the environment, the first path of motion comprising a first motion and a second motion;
determining a cost of the first motion using the costmap;
determining a cost of the second motion using the costmap; and
determining a total cost of the first path of motion.

9. The method of claim 8 , further comprising:
determining a total cost of a second path of motion;
determining that the total cost of the second path of motion is the low cost; and
selecting the second path of motion.

10. The method of claim 2 , further comprising:
determining a cost for the movement through the environment for the navigation using the costmap and the swept volume.

11. The method of claim 2 , wherein detecting the change comprises detecting the change using one or more cameras.

12. A method, comprising:
transmitting, by a server, a map of an environment to a first autonomous cart, the map comprising a costmap that represents a cost to the first autonomous cart for moving through a certain portion of the environment, wherein the first autonomous cart uses the costmap to navigate through the environment at a low cost, and wherein the first autonomous cart is configured to determine a swept volume that represents a sum of footprints that the first autonomous cart will occupy after navigating through the environment;
receiving, by the server, a representation of a change in the environment detected by the first autonomous cart, the change being indicative of at least one of an alteration in visual appearance of the environment with respect to the map and a discovery of a new portion of the environment not represented in the map;
generating, by the server, an updated map of the environment, the updated map showing the change detected by the first autonomous cart; and
transmitting, by the server, the updated map to at least one of the first autonomous cart or a second autonomous cart.

13. A system, comprising:
a server; and
a first autonomous cart, the first autonomous cart communicatively coupled to the server, the first autonomous cart configured to:
obtain a map of an environment from the server, the map comprising a costmap that represents a cost to the first autonomous cart for moving through a certain portion of the environment;
determine a swept volume for the first autonomous cart, wherein the swept volume represents a sum of footprints that the first autonomous cart will occupy after navigating through the environment;
navigate through the environment based on the costmap at a low cost, wherein the cost is determined using the costmap;
detect a change in the environment with respect to the map while navigating through the environment, the change being at least one of an alteration in visual appearance of the environment with respect to the map or a discovery of a new portion of the environment not represented in the map; and
transmit a representation of the change to the sever.

14. The system of claim 13 , wherein the first autonomous cart is further configured to:
obtain an updated map of the environment from the server, the updated map showing the change detected by the first autonomous cart.

15. The system of claim 13 , wherein the costmap further represents at least one of a passable area or an impassable area for the first autonomous cart in the environment.

16. The system of claim 15 , wherein the first autonomous cart is further configured to:
generate a two-dimensional (2D) traversability map based on the costmap, the 2D traversability map including traversability information for the first autonomous cart.

17. The system of claim 16 , wherein the first aut

Assignees

• A9 Com Inc

Inventors

• Keivan Nima
• Falquez Juan
• Hoffman Andrew E
• Mcqueen Jonathan

Classifications

• G05D1/0274

  using mapping information stored in a memory device (navigation using map-matching G01C21/30) · CPC title

• G05D1/0214 Primary

  in accordance with safety or protection criteria, e.g. avoiding hazardous areas (monitoring the location of vehicles within a certain area, e.g. forbidden or allowed areas, in traffic control systems for road vehicles G08G1/13) · CPC title

• G05D1/0217

  in accordance with energy consumption, time reduction or distance reduction criteria · CPC title

• G05D1/0289

  with means for avoiding collisions between vehicles (vehicle fittings for automatically controlling speed including means for detecting potential obstacles B60K31/0008; avoiding obstacles by action on the steering system B62D; radar, sonar, lidar systems designed for anti-collision purposes G01S13/93, G01S15/93, G01S17/93) · CPC title

• G05D1/0246

  using a video camera in combination with image processing means · CPC title