Peace!
ロボット使ってトマト作れば楽だなぁ...やってみっか!
農業ロボット開発記録
日本中で100台以上動かしちゃった(^^)/
I'm developing robots that do tomato harvesting, settled planting ,,,,and many tasks.
The goal of production cost of robot is $2,000 per unit.
Because it is necessary a lot of robots, in order to innovate the farming.
I use Raspberry Pi as robot controller to develop low cost robots.
TOMATO FARM website -> CLICK!
2012年10月31日水曜日
I'm a farmer. v(^o^)/ cultivate plants using cell tray
2012年10月30日火曜日
Design a motion of 4 legged robot
I have designed a motion of 4 legged robot using following document.
This document describes a motion pattern and how to move each legs, heels and body.
I have been testing effective walking of 4 legged robot for a half month.
I have gathered data and wrote the document. And I have coded the walking program of 4 legged robot.
Peace!
New house & new machine
Kurin is happy.
Because he has a new house.
Now Japan is autumn, It will be winter soon.
I wonder if his house will warm KURIN in winter.
A new cutting machine!
I use this machine to make a robot part.
It can cut materiales at a right angle.
To cut materiales at a right angle is very important to build up robots.
Peace!
Because he has a new house.
Now Japan is autumn, It will be winter soon.
I wonder if his house will warm KURIN in winter.
A new cutting machine!
I use this machine to make a robot part.
It can cut materiales at a right angle.
To cut materiales at a right angle is very important to build up robots.
Hitachi Koki FC7FSB |
Peace!
2012年10月29日月曜日
The Video !! Walking a 4 legged robot. v(^.^)v
Please see a prototype robot video.
It's a powerful 4 legged robot.
This prototype robot is used for functional verification.
It seem a simple robot, but it have many function.
Space recognition, obstacle detection, autonomous locomotion,,,,
These function is on my original robot platform called 'RDCS'.
I will improve these robot function using results of prototype robot testing.
By the way, he recognizes his environment through his eye (xtion).
Peace!
It's a powerful 4 legged robot.
This prototype robot is used for functional verification.
It seem a simple robot, but it have many function.
Space recognition, obstacle detection, autonomous locomotion,,,,
I will improve these robot function using results of prototype robot testing.
By the way, he recognizes his environment through his eye (xtion).
Peace!
2012年10月27日土曜日
I'm a farmer. v(^o^)/
2012年10月21日日曜日
I'm a farmer. (o^_^o)...
I seeded today.
Onion, broccoli, daikon radish, , ,
and tomatoes for testing of harvesting robot.
happy!
Onion, broccoli, daikon radish, , ,
and tomatoes for testing of harvesting robot.
happy!
KONG LIVES!
Do you know this badge?
It was about 20 years ago. This badge passed out after an incident at Universal Studios Hollywood.
peace!
It was about 20 years ago. This badge passed out after an incident at Universal Studios Hollywood.
peace!
2012年10月19日金曜日
Useful robot work bench.
I made a robot work bench yesterday. I could understand it is very useful.
The photo is a scene of test recognizing target.
I can move a Xbox on bench smoothly and easily.
Peace!
Now status:4 legged robot. New robot bench.
I'm developing a motion realtime generation program for my 4 legged robot.
It's very difficult program for me. I'd like to develop the program in this week.
I made a robot bench. When I debug a motion realtime generation program, i use this bench. It took 4 hours to make this bench.
Now I develop a prototype 4 legged robot. This bench is able to use too, when i will develop a next robot which work in my farm and greenhouse.
prototype 4 legged robot |
prototype 4 legged robot (front) |
robot bench |
robot bench |
peace!
2012年10月12日金曜日
A new face
I've developed a forward obstacle detection system of my prototype robot.
When there are obstacles that can not be over in the range of 30cm front, my robot avoid the obstacles.
ロボットの前面に、障害回避の機能を設置しました。
ロボットの前、30cm程度に存在する、乗り越えられない障害物を回避します。
Agricultural robot will be operated in outdoor such as 'farm' and 'greenhouse'. Therefore, it is more complex than the robot to work indoors. It require dustproof, waterproof,,,. However, the production cost is $ 2,000. It is a very difficult challenge.
I developed a 1/2 scale prototype robot , and I'm testing the robot function.
There are many improvements and issues. (~.~), foo
I have to resolve and advance step by step.
農業ロボットは、農場やハウスなどの野外で運用されます。室内で活動するロボットより複雑です。
例えば、防塵、防滴などが必要になります。それも、2000ドルで開発しなければなりません。難しい挑戦です。1/2スケールの試作機を作りながら、プログラムなどの最終チェックをしています。
I didn't describe how to recognize tomatoes, because it's a new method.
I will explain, when the test of recognizing tomatoes will finish.
トマトの認識技術については、これまで書きませんでした。
従来のように、画像処理を中心にした認識技術では有りません。テストが終わった時に説明します。
peace!
When there are obstacles that can not be over in the range of 30cm front, my robot avoid the obstacles.
ロボットの前面に、障害回避の機能を設置しました。
ロボットの前、30cm程度に存在する、乗り越えられない障害物を回避します。
Agricultural robot will be operated in outdoor such as 'farm' and 'greenhouse'. Therefore, it is more complex than the robot to work indoors. It require dustproof, waterproof,,,. However, the production cost is $ 2,000. It is a very difficult challenge.
I developed a 1/2 scale prototype robot , and I'm testing the robot function.
There are many improvements and issues. (~.~), foo
I have to resolve and advance step by step.
農業ロボットは、農場やハウスなどの野外で運用されます。室内で活動するロボットより複雑です。
例えば、防塵、防滴などが必要になります。それも、2000ドルで開発しなければなりません。難しい挑戦です。1/2スケールの試作機を作りながら、プログラムなどの最終チェックをしています。
I didn't describe how to recognize tomatoes, because it's a new method.
I will explain, when the test of recognizing tomatoes will finish.
トマトの認識技術については、これまで書きませんでした。
従来のように、画像処理を中心にした認識技術では有りません。テストが終わった時に説明します。
peace!
2012年10月11日木曜日
Nexus 7 broken! (-.-)...
Today Suddenly, Nexus 7 don't boot.
Google service said "It's initial failure."
Over one week, I can't develop a image processing software on Nexus 7.
It's shocked.
我慢
Google service said "It's initial failure."
Over one week, I can't develop a image processing software on Nexus 7.
It's shocked.
我慢
2012年10月8日月曜日
A new robot on 4 legs. (prototype)
Now, I'm developing a 4 legs robot. The pictures are a prototype 4 legs robot.
The prototype robot is half size model.
4本脚ロボットを開発しています。写真は、そのプロトタイプです。
I grow organic tomatoes in the greenhouse. And in the greenhouse, an aisle is narrow (40cm) and uneven ground.
There are several robot moving methods that are wheels, caterpillar, 4 legs and others. I have considered which is the best moving method in the 'farm' and 'greenhouse'.
I decided to develop a 4 legs robot, first of all.
私は、化学農薬などを使わないオーガニック トマトを農業ハウスで栽培しています。
ハウス内は、通路も狭く、また平ではありません。
ロボットの移動方法には、車輪、キャタピラー、4本脚、それ以外の方法もあります。
最初に、4本脚のロボットを作る事に決めました。
The selection points are below:
(1) A function of holding on the level ( a height level requirement )
My forming robot have to move on the application which are an actuator, pruner machine,,,
Therefore a robot need to hold these applications on the level in order to exercise their capacities.
A wheel and caterpillar machin are difficult to hold on the level.
But 4 legs moving system is easy to realize my height level requirement, nothing any other parts.
I'm going to research and to develop for putting into practical use.
peace!
The prototype robot is half size model.
4本脚ロボットを開発しています。写真は、そのプロトタイプです。
30cm ruler in lower side |
I grow organic tomatoes in the greenhouse. And in the greenhouse, an aisle is narrow (40cm) and uneven ground.
There are several robot moving methods that are wheels, caterpillar, 4 legs and others. I have considered which is the best moving method in the 'farm' and 'greenhouse'.
I decided to develop a 4 legs robot, first of all.
私は、化学農薬などを使わないオーガニック トマトを農業ハウスで栽培しています。
ハウス内は、通路も狭く、また平ではありません。
ロボットの移動方法には、車輪、キャタピラー、4本脚、それ以外の方法もあります。
最初に、4本脚のロボットを作る事に決めました。
The selection points are below:
(1) A function of holding on the level ( a height level requirement )
My forming robot have to move on the application which are an actuator, pruner machine,,,
Therefore a robot need to hold these applications on the level in order to exercise their capacities.
A wheel and caterpillar machin are difficult to hold on the level.
But 4 legs moving system is easy to realize my height level requirement, nothing any other parts.
I'm going to research and to develop for putting into practical use.
peace!
I'm a farmer. (o^_^o)...
It's my tractor.
peace!
It's 20 horsepower.
Back of the tractor, greenhouse of tomatos harvesting. My tomatos are organic productions.
Next year, robot will work in greenhouse of tomatos harvesting.
Next year, robot will work in greenhouse of tomatos harvesting.
peace!
2012年10月5日金曜日
2012年10月3日水曜日
Performance testing Xtion on Raspberry Pi
Using a capturing server
Raspberry Pi is a good machine. But it don't have a capacity for 3D processing. For this test, I use a Raspberry Pi as a capturing service (server). And MacMini is 3D Processing service and viewing.
My test environment diagram is below.
Raspberry Piは、3D処理をするには性能がたりません。従って、Xtionを使ったcaputuring setverとして使い、空間認識などの処理をMacminiで処理します。そのテストをしました。
テストは、RDCS環境を使いまいた。RDCSは、分散コンピューティングをベースにしたロボットプラットフォームです。その機能を使ってテストをしました。
Micminiでpoint cloud処理をしています。さすがに瞬間的な処理です。
Capturing Service 3D recognizing service
Xtion
Raspberry Pi Mac MIni
---------------------------------------------------- Coordinator Layer
---------------------------------------------------- (Service Control)
| |
---------------------------------------------------- Cooperative Layer
---------------------------------------------------- (Communication Control)
A performance testing was executed on RDCS.
RDCS is a robot platform based on a distributed computing model.
A lots of robot function (hardware, software) is connected by "Cooperative Control" and "Coordinator".
"Cooperative Control" is a network communication control. The control connects computer , actuator, sensor and other robot device.
"Coordinator" offers robot required software. "Coordinator" integrates a lot robot software on RDCS.
This test flow is :
(1) 3D recognized service requests a data set ( image and depth ) to Capturing service.
(2) Capturing service on Raspberry Pi execute to capture image and depth.
(3) 3D recognized service receive a couple of capturing data.
(4) 3D processing and viewing
RESULT (sec)
(1) 0-1
(2) 8-10 (include the time which is Xtion initialize and terminate)
A Capture time is 1-2 sec.
(3) 0-1
(4) 0-1
This result means that a single robot don't have all function in it.
Therefore, it's able to develop a lightweight and low cost robot.
There are a lot of kinds wrok in farm. And my farm is large. Thus I need a lot of robot. To do that, I have to develop a low cost and multi-function robot.
My robots are structured by base robot and add-in function in order to realize a multi-function of robot.
And these robots have to been set up by commercialized product and builded by component-based.
Now I'm developing a core component of robot using 2 or 3 Raspberry Pi.
A core component of robot is structured by 6 components which are robot platform(RDCS), space recognition, obstacle detection, moving, fail-safe and recovery.
An effectivity of RDCS which I have developed was confirmed by this testing.
この結果は、1台のロボットに全ての機能を集約しなくても良い事を示しています。
従って、軽量なかつ低価格のロボットを作る事ができます。
ロボットプラットフォーム、障害物検知、通信、移動、フェイルセーフと回復機能を持ったシンプルなロボットを、Raspberry Piを2、3個使って作れば良いと思います。その上に、アクチュエータなどのロボットのパーツを載せればよのではないでしょうか?
私の開発している、RDCSのアーキテクチャーの有効性を確認できました。
3D point cloud
Peace!
My test environment diagram is below.
Raspberry Piは、3D処理をするには性能がたりません。従って、Xtionを使ったcaputuring setverとして使い、空間認識などの処理をMacminiで処理します。そのテストをしました。
テストは、RDCS環境を使いまいた。RDCSは、分散コンピューティングをベースにしたロボットプラットフォームです。その機能を使ってテストをしました。
Micminiでpoint cloud処理をしています。さすがに瞬間的な処理です。
Capturing Service 3D recognizing service
Xtion
Raspberry Pi Mac MIni
---------------------------------------------------- Coordinator Layer
---------------------------------------------------- (Service Control)
| |
---------------------------------------------------- Cooperative Layer
---------------------------------------------------- (Communication Control)
A performance testing was executed on RDCS.
RDCS is a robot platform based on a distributed computing model.
A lots of robot function (hardware, software) is connected by "Cooperative Control" and "Coordinator".
"Cooperative Control" is a network communication control. The control connects computer , actuator, sensor and other robot device.
"Coordinator" offers robot required software. "Coordinator" integrates a lot robot software on RDCS.
This test flow is :
(1) 3D recognized service requests a data set ( image and depth ) to Capturing service.
(2) Capturing service on Raspberry Pi execute to capture image and depth.
(3) 3D recognized service receive a couple of capturing data.
(4) 3D processing and viewing
RESULT (sec)
(1) 0-1
(2) 8-10 (include the time which is Xtion initialize and terminate)
A Capture time is 1-2 sec.
(3) 0-1
(4) 0-1
This result means that a single robot don't have all function in it.
Therefore, it's able to develop a lightweight and low cost robot.
There are a lot of kinds wrok in farm. And my farm is large. Thus I need a lot of robot. To do that, I have to develop a low cost and multi-function robot.
My robots are structured by base robot and add-in function in order to realize a multi-function of robot.
And these robots have to been set up by commercialized product and builded by component-based.
Now I'm developing a core component of robot using 2 or 3 Raspberry Pi.
A core component of robot is structured by 6 components which are robot platform(RDCS), space recognition, obstacle detection, moving, fail-safe and recovery.
An effectivity of RDCS which I have developed was confirmed by this testing.
この結果は、1台のロボットに全ての機能を集約しなくても良い事を示しています。
従って、軽量なかつ低価格のロボットを作る事ができます。
ロボットプラットフォーム、障害物検知、通信、移動、フェイルセーフと回復機能を持ったシンプルなロボットを、Raspberry Piを2、3個使って作れば良いと思います。その上に、アクチュエータなどのロボットのパーツを載せればよのではないでしょうか?
私の開発している、RDCSのアーキテクチャーの有効性を確認できました。
3D point cloud
Peace!
登録:
投稿 (Atom)