Television – Special applications – Navigation
Reexamination Certificate
2000-02-28
2004-02-03
Le, Vu (Department: 2613)
Television
Special applications
Navigation
C348S120000
Reexamination Certificate
active
06686951
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates generally to systems that aid in crop cultivation. More particularly, the invention relates to an automated steering control system for a robotic cultivation system that utilizes an electronic vision guidance system.
BACKGROUND OF THE INVENTION
Guidance systems have been proposed for controlling the movement of vehicles traveling along predetermined paths such as, for example, robotic material handling guidance systems. In some applications, robotic vehicles for moving materials between separate points in a warehouse or storage area without human intervention, are provided with automated steering control systems that utilize inductive sensors mounted near the bottom of the vehicle in order to sense currents passing through thin wires disposed either at or below the surface of predetermined pathways. The inductive sensors sense the magnetic field radiated from the wire and, using conventional phase detection techniques, produce an output signal which indicates the extent to which a vehicle has drifted laterally from the desired course. The output signals from the sensors are provided to a vehicle control circuit that typically converts the sensor output signals into control error signals which are used to direct a vehicle back on course.
These inductive types of lateral guidance control systems have experienced some success in relatively slow-moving, industrial materials handling applications. However, these systems are considered to have serious drawbacks when they are contemplated for controlling motor vehicles moving at appreciable speeds on a highway or for controlling cultivating and agricultural vehicles in a farm field. For example, the magnetic field radiated by a current conducting wire disposed in the surface of the road or farm field may be distorted by metallic structures along the roadway or by stray magnetic fields from nearby power lines. Such inductive sensors are highly prone to sensing spurious signals radiated by other electromagnetic sources. In addition, such a system requires the installation and maintenance of power cables, transformers and power supplies along the road or along the farm field, which adds an appreciable cost to such a system. Furthermore, the performance of such a system is fairly limited because an inductive sensor cannot “look-ahead” as a human driver does. Inductive sensor systems can only react to the magnetic field received from the roadway or farm field immediately beneath the vehicle. Consequently, without a “look-ahead” capability, the reaction times of the inductive sensor control systems are very slow in comparison with those of a driver.
Considering lateral guidance control systems applied specifically to crop harvesting machinery or other cultivation machinery, a number of limitations are apparent. The farm field is not conducive to having current-conducting wires disposed beneath the earth's surface because cultivation may cause damage to such wires. Further, maintenance of such power cables is intractable for farmers who would find maintenance to require expensive equipment or expensive servicing. Further still, the current conducting wires are not likely to be well aligned with rows of crops in the farm field.
The task of manually steering multiple row harvesting machines, for example moving over a farm field along rows at a modest rate of speed can be a tiresome task. In harvesting equipment, the row crop separators which guide the crop to the cutting elements of the machine are often obscured by dust, crop material, and/or vegetation such that precise manual steering by visual observation is difficult if not impossible. To alleviate this problem, alternative steering control systems have been implemented which use mechanical feelers to determine the position of standing row crop relative to the row crop separators. While steering control systems with mechanical feelers offer an improvement over guidance by visual observation, or by current conducting wires, these systems inherently have certain disadvantages associated therewith. For example, the mechanical feelers are subject to fouling and damage. In addition, because of the rough and dirty environment in which the feelers must operate, they must be serviced frequently to insure that they are mechanically free. Furthermore, for spraying operations, high speeds are desirable to enable a ground vehicle to challenge the role of a crop-spraying aircraft. Because of the difficulty in controlling a crop sprayer at a high rate of speed over the farm field, the driver's task is considerably demanding or intractable and therefore an automated guidance system is desirable.
Vision guidance systems have been designed to steer a tractor relative to the rows of a crop in a farm field. However, because of the odd geometry of farm crops and the ever changing geometry of farm crops, due to growth and weed growth, the vision guidance system must be insensitive to visual noise, while also adapting to loss of growth in, for example, a barren patch of the field.
Thus, there is a need for a vision based guidance system that is insensitive to visual noise. There is also a need for a vision based guidance system for a crop cultivation system that is robust in barren patches of the field. Further, there is a need for a vision based guidance system that utilizes algorithms for rapid data processing. Further still, there is a need for a vision based guidance system for an agricultural vehicle that includes a trajectory planner capable of “looking-ahead” in the farm field. Further still, there is a need for a vision based guidance system for an agricultural vehicle that utilizes regions of interest to identify vegetation based upon a cost optimization. Even further still, there is a need for a vision based guidance system for an agricultural vehicle that utilizes predetermined distances between rows to aid in guidance of the system. Still even further, there is a need for a vision based guidance system that utilizes a multiplicity of rows for tracking.
SUMMARY OF THE INVENTION
One embodiment of the invention relates to an automatic guidance system for an agricultural vehicle. The guidance system includes an electromagnetic sensor for gathering information about field characteristics. The electromagnetic sensor is configured to provide a digital signal representative of the field characteristics. The digital signal is representative of pixels of an image. The invention also relates to an information processor having a memory and a central processing unit. The information processor is coupled to the electromagnetic sensor. Further, the invention relates to a program in the memory of the information processor configured to be executed by the central processing unit. The program is configured to categorize the pixels into categories of field characteristics.
Another embodiment of the invention relates to a method of sensing field characteristics. The method includes sensing electromagnetic radiation with an electromagnetic sensor. The method also includes communicating a digital signal representative of the sensed electromagnetic radiation, and representative of pixels of an image, to a processing device. Further, the method includes categorizing the pixels into field characteristic categories by using a clustering operation.
Yet another embodiment of the invention relates to an agricultural vehicle configured to be guided through a field by an automatic guidance system. The agricultural vehicle includes a visual means for gathering visual input about field characteristics in the format of a digitized image, including pixel information. The invention also includes a clustering means configured to cluster pixel information into groups.
REFERENCES:
patent: 3834036 (1974-09-01), Scarritt, Sr.
patent: 4165796 (1979-08-01), Petz
patent: 4211921 (1980-07-01), Kanetou et al.
patent: 4298084 (1981-11-01), Newell
patent: 4366756 (1983-01-01), Brum
patent: 4505094 (1985-03-01), Demorest
patent: 4515221 (1985-05-01), van der Lely
patent: 4528804 (
Dickson Monte Andre
Han Shufeng
Ni Bingcheng
Case LLC
Henkel Rebecca
Le Vu
Maurer Brant T.
Senfi Behrooz
LandOfFree
Crop row segmentation by K-means clustering for a vision... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Crop row segmentation by K-means clustering for a vision..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Crop row segmentation by K-means clustering for a vision... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3351108