TWI806721B - Autonomous Mobile Artificial Intelligence Mushroom Cultivation Monitoring System and Method - Google Patents
Autonomous Mobile Artificial Intelligence Mushroom Cultivation Monitoring System and Method Download PDFInfo
- Publication number
- TWI806721B TWI806721B TW111128426A TW111128426A TWI806721B TW I806721 B TWI806721 B TW I806721B TW 111128426 A TW111128426 A TW 111128426A TW 111128426 A TW111128426 A TW 111128426A TW I806721 B TWI806721 B TW I806721B
- Authority
- TW
- Taiwan
- Prior art keywords
- mushroom
- module
- growth
- growth environment
- image
- Prior art date
Links
Images
Landscapes
- Mobile Radio Communication Systems (AREA)
- Cultivation Of Plants (AREA)
Abstract
本發明揭露一種自主移動式人工智慧菇類栽培監控系統及方法,其包括單兵式自主移動巡檢載具、影像擷取單元、生長環境感測單元、場域圖資建置模組、生長狀態辨識模組、巡檢路徑規劃模組及生長環境調控模組。以影像擷取單元擷取菇類栽培場域內的地圖影像及菇類影像。以場域圖資建置模組依據地圖影像建置地面通道圖資及菇籃置架圖資。以生長環境感測單元感測菇類之生長環境數據。以生長狀態辨識模組將菇類影像與基準菇類影像比對而判斷菇類是否為遲緩生長。以巡檢路徑規劃模組依據地面通道圖資而規劃設定行走路徑及位在行走路徑上的複數個停駐位置,及依據菇籃置架圖資而規劃設定檢測路徑及位在檢測路徑上的複數個檢測位置。以行走控制模組控制單兵式自主移動巡檢載具依據行走路徑而在地面通道行走及於每一停駐位置停駐。影像擷取單元及生長環境感測單元隨著單兵式自主移動巡檢載具及移載機構而依序移動至每一停駐位置及檢測位置,以擷取菇類影像及感測生長環境數據。當生長狀態辨識模組判斷菇類為遲緩生長時,生長環境調控模組依據資料庫中預設而相符合的一第一生長環境條件參數來調控該菇類栽培場域內的溫度、濕度及二氧化碳濃度,以加速菇類生長。藉以達到單兵作業以簡化系統且能智慧有效地對於菇類栽培場域進行生長監控之目的。 The present invention discloses an autonomously mobile artificial intelligence mushroom cultivation monitoring system and method, which includes a single-soldier autonomously mobile inspection vehicle, an image capture unit, a growth environment sensing unit, a field map information building module, a growth State identification module, inspection path planning module and growth environment control module. The map image and the mushroom image in the mushroom cultivation field are captured by the image capture unit. Use the field map data building module to build ground channel map data and mushroom basket rack map data based on map images. The growth environment data of mushrooms is sensed by the growth environment sensing unit. The growth state recognition module compares the mushroom image with the reference mushroom image to determine whether the mushroom is growing slowly. Use the inspection path planning module to plan and set the walking path and multiple parking positions on the walking path according to the map information of the ground passage, and plan and set the detection path and the parking positions on the detection path according to the map information of the mushroom basket rack Multiple detection positions. Use the walking control module to control the individual autonomous mobile inspection vehicle to walk on the ground passage and park at each parking position according to the walking path. The image capture unit and growth environment sensing unit move to each parking position and detection position sequentially with the individual autonomous mobile inspection vehicle and transfer mechanism to capture mushroom images and sense the growth environment data. When the growth state identification module judges that the mushroom grows slowly, the growth environment regulation module regulates the temperature, humidity and Carbon dioxide concentration to accelerate the growth of mushrooms. In order to achieve the purpose of individual operation to simplify the system and intelligently and effectively monitor the growth of mushroom cultivation fields.
Description
本發明係有關一種自主移動式人工智慧菇類栽培監控系統及方法,尤指一種可藉由可視化生長環境圖表之機能設置以供快速地解讀出菇類生長環境資訊的人工智慧菇類栽培監控技術。 The present invention relates to a self-moving artificial intelligence mushroom cultivation monitoring system and method, especially an artificial intelligence mushroom cultivation monitoring technology that can quickly interpret the mushroom growth environment information through the function setting of the visual growth environment chart .
按,在台灣的食用菇產業從1909年起有了香菇段木人為栽種紀錄,截至目前已經有百年以上的發展歷史,而在國內種植之各式各樣的菇類當中,杏鮑菇在台灣起源於1996年,然而在短短十年內,杏鮑菇就已經取代金針菇成為產值第二高的菇種,於是不難發現杏鮑菇是唯一在近年產量仍保持成長且價格波動小之菇種。目前杏鮑菇與金針菇同樣是環境控制栽培的菇種,在自動化程度卻遠遠不及金針菇,原因包含一般業者是多數是以袋栽太空包進行生產,而非瓶栽式,其次是杏鮑菇相較其他菇種,自動化開發起步較晚,特別在採收流程這塊,而袋栽太空包的輪廓、尺寸相較於瓶栽的輪廓尺寸一致性低,以及台灣菌種商在國內數量佔比較少,主要以製包業者及栽培業者占大多數,凸顯出菌種少的問題,而這點也間接影響到杏鮑菇最終生成的外觀上,對於採收自動化有一定程度的影響。 By the way, the edible mushroom industry in Taiwan has a record of man-made planting of shiitake mushrooms since 1909. Up to now, it has a development history of more than one hundred years. Among the various mushrooms grown in China, Pleurotus eryngii is the most popular in Taiwan It originated in 1996, but in just ten years, Pleurotus eryngii has replaced Flammulina velutipes as the second highest value mushroom species, so it is not difficult to find that Pleurotus eryngii is the only mushroom whose output has maintained growth in recent years and has little price fluctuation kind. At present, Pleurotus eryngii and Flammulina velutipes are also cultivated under environmental control, but the degree of automation is far inferior to Flammulina velutipes. The reasons include that most of the industry is produced in bag-grown space packs instead of bottle-grown, followed by Pleurotus eryngii Compared with other mushroom species, automation development started late, especially in the harvesting process, while the outline and size of bag-grown space packs are less consistent than those of bottle-grown, and the number of Taiwan-grown strain suppliers in China accounts for Relatively few, mainly packagers and cultivators account for the majority, highlighting the problem of fewer strains, and this also indirectly affects the final appearance of Pleurotus eryngii, which has a certain degree of impact on harvesting automation.
此外,傳統的菇類產業,需要依靠大量的人力進行巡檢工作,庫房內的環境狹窄潮濕且庫房的數量眾多,若以傳統人力來進行巡檢工作,會耗費大量時間與人員配置;另外菇類於成熟時所擴散 的孢子,也會對菇農的呼吸道造成影響。若能將智慧農業應用在菇類產業上,將可大幅減低依賴人力的需求,並達到自動化生產的目標。 In addition, the traditional mushroom industry requires a lot of manpower for inspection work. The environment in the warehouse is narrow and humid and there are a large number of warehouses. If traditional manpower is used for inspection work, it will take a lot of time and staffing; similar to that diffused at maturity The spores will also affect the respiratory tract of mushroom growers. If smart agriculture can be applied to the mushroom industry, it will greatly reduce the need to rely on manpower and achieve the goal of automated production.
為實現自動化巡檢之目的,於是本申請人乃提出一件如發明證書第I719924號『菇類生長監控系統與方法』所示的專利,其包括巡檢自駕車、生長環境調控模組及中央控制單元。巡檢自駕車包括具有行走輪的動力驅行裝置、訊號處理單元、生長環境狀態感測模組及生長影像擷取單元。訊號處理單元控制巡檢自駕車沿著佈設在菇類栽培場域的地面通道上之預設巡檢路徑行走,依序抵達分佈在預設巡檢路徑上的巡檢停駐位置及停駐位置,並控制生長環境狀態感測模組及生長影像擷取單元分別感測即時菇類的生長環境狀態及菇類生長影像。中央控制單元接收處理菇類的生長環境狀態及菇類生長影像而分別於相應的即時栽培時間序列產生即時生長環境狀態參數及即時生長影像特徵參數,並將即時生長環境狀態參數、即時生長影像特徵參數與相應的基準菇類栽培時間序列之基準生長環境狀態參數及基準生長影像特徵參數比對,當比對結果之差異超過一預定範圍時,該中央控制單元控制啟動生長環境調控設備,以調節該菇類栽培場域的菇類生長環境狀態。 In order to achieve the purpose of automatic inspection, the applicant proposed a patent as shown in the invention certificate No. I719924 "Mushroom Growth Monitoring System and Method", which includes inspection self-driving cars, growth environment control modules and central control unit. The inspection self-driving car includes a power drive device with traveling wheels, a signal processing unit, a growth environment state sensing module, and a growth image capture unit. The signal processing unit controls the inspection self-driving car to walk along the preset inspection path arranged on the ground passage of the mushroom cultivation field, and arrive at the inspection parking positions and parking positions distributed on the preset inspection path in sequence , and control the growth environment state sensing module and the growth image acquisition unit to sense the real-time mushroom growth environment state and the mushroom growth image respectively. The central control unit receives and processes the growth environment status and mushroom growth images of mushrooms to generate real-time growth environment status parameters and real-time growth image feature parameters in the corresponding real-time cultivation time series, and generates real-time growth environment status parameters and real-time growth image feature parameters. The parameters are compared with the reference growth environment state parameters and reference growth image characteristic parameters of the corresponding reference mushroom cultivation time series. When the difference between the comparison results exceeds a predetermined range, the central control unit controls and starts the growth environment control equipment to adjust The state of the mushroom growth environment in the mushroom cultivation field.
該專利雖然具備菇類栽培場域自動巡檢及依據生長狀態調節菇類生長環境等功能;惟,該專利並無可視化生長環境圖表的功能設置,以致無法讓操作人員可以快速地解讀出菇類生長環境資訊,致使較難以營建出更為適合菇類栽種生長的環境;不僅如此,該專利無法利用大數據收集、人工智慧框選及訓練,以致無法對庫房內的菇類進行生產數量的預測,致使較無法因應市場需求而加快或減緩菇類的生長速度。 Although the patent has functions such as automatic inspection of the mushroom cultivation field and adjustment of the mushroom growth environment according to the growth state; however, the patent does not have the function setting of a visual growth environment chart, so that the operator cannot quickly interpret the mushrooms. The growth environment information makes it more difficult to create a more suitable environment for the growth of mushrooms; moreover, this patent cannot use big data collection, artificial intelligence frame selection and training, so that it is impossible to predict the production quantity of mushrooms in the warehouse , making it impossible to accelerate or slow down the growth rate of mushrooms in response to market demand.
因鑑於目前尚無一種無可視化生長環境圖表及菇類產量預測的菇類栽培監控技術、專利或是論文的公開或是發表,因此,上述習知技術及前述專利確實皆未臻完善,仍然有再改善的必要性;緣是,本發明人乃積極投入研發,終而有本發明的研發成果產出。 In view of the fact that there is no mushroom cultivation monitoring technology, patent, or paper published or published without visual growth environment charts and mushroom yield predictions, the above-mentioned conventional technologies and the aforementioned patents are indeed not perfect, and there are still The necessity of further improvement; the reason is that the inventor is actively investing in research and development, and finally has the research and development results of the present invention.
本發明第一目的,在於提供一種單兵作業以簡化系統且能智慧有效地對於菇類栽培場域進行生長監控的自主移動式人工智慧菇類栽培監控系統及方法。達成本發明第一目的所採用之技術手段,係包括單兵式自主移動巡檢載具、影像擷取單元、生長環境感測單元、場域圖資建置模組、生長狀態辨識模組、巡檢路徑規劃模組、生長環境調控模組及中央處理單元。以該影像擷取單元擷取菇類栽培場域內的地圖影像及菇類影像,並以場域圖資建置模組依據地圖影像建置包括有菇類栽培場域的地面通道圖資及菇籃置架圖資之場域地圖圖資。以生長環境感測單元感測菇類栽培場域內的菇類之生長環境數據。以生長狀態辨識模組將菇類影像與一資料庫中預設的基準菇類影像比對,而判斷菇類是否為遲緩生長。以巡檢路徑規劃模組依據地面通道圖資而規劃設定行走路徑及位在行走路徑上的複數個停駐位置,及依據菇籃置架圖資而規劃設定檢測路徑及位在檢測路徑上的複數個檢測位置。以行走控制模組控制單兵式自主移動巡檢載具依據行走路徑而在菇類栽培場域的地面通道行走及依序於每一停駐位置停駐。影像擷取單元及生長環境感測單元經由移載機構而設於單兵式自主移動巡檢載具上,並隨著該單兵式自主移動巡檢載具依序移動至每一停駐位置,並由移載機構驅使影像擷取單 元及生長環境感測單元依序於每一檢測位置停駐,使影像擷取單元得以擷取菇類影像及生長環境感測單元得以感測生長環境數據。以生長環境調控模組驅動生長環境調控設備以調控該菇類栽培場域內的溫度、濕度及二氧化碳濃度。當生長狀態辨識模組判斷該至少一菇類為遲緩生長時,該生長環境調控模組則依據該資料庫中預設而相符合的一第一生長環境條件參數來調控該菇類栽培場域內的溫度、濕度及二氧化碳濃度,以加速菇類生長。 The first purpose of the present invention is to provide an autonomous mobile artificial intelligence mushroom cultivation monitoring system and method that can simplify the system and intelligently and effectively monitor the growth of the mushroom cultivation field. The technical means adopted to achieve the first objective of the present invention include a single-soldier autonomous mobile inspection vehicle, an image capture unit, a growth environment sensing unit, a field map data construction module, a growth status identification module, Inspection path planning module, growth environment regulation module and central processing unit. Use the image capture unit to capture map images and mushroom images in the mushroom cultivation field, and use the field map information construction module to construct ground passage map information including the mushroom cultivation field and Field map map information for mushroom basket racks. The growth environment data of the mushrooms in the mushroom cultivation field is sensed by the growth environment sensing unit. The growth state recognition module compares the mushroom image with the preset reference mushroom image in a database, and judges whether the mushroom grows slowly. Use the inspection path planning module to plan and set the walking path and multiple parking positions on the walking path according to the map information of the ground passage, and plan and set the detection path and the parking positions on the detection path according to the map information of the mushroom basket rack Multiple detection positions. Use the walking control module to control the individual autonomous mobile inspection vehicle to walk on the ground passage of the mushroom cultivation field according to the walking path and stop at each parking position in sequence. The image capture unit and the growth environment sensing unit are installed on the individual autonomous mobile inspection vehicle through the transfer mechanism, and move to each parking position sequentially with the individual autonomous mobile inspection vehicle , and the image capture unit is driven by the transfer mechanism The element and the growth environment sensing unit stop at each detection position sequentially, so that the image capture unit can capture mushroom images and the growth environment sensing unit can sense growth environment data. The growth environment control device is driven by the growth environment control module to control the temperature, humidity and carbon dioxide concentration in the mushroom cultivation field. When the growth state identification module judges that the at least one mushroom grows slowly, the growth environment regulation module regulates the mushroom cultivation field according to a first growth environment condition parameter preset and matched in the database The temperature, humidity and carbon dioxide concentration inside to accelerate the growth of mushrooms.
本發明第二目的,在於提供一種具備菇類產量預測功能的自主移動式人工智慧菇類栽培監控系統及方法,主要是可以利用大數據收集、人工智慧框選及訓練,以對庫房內的菇類進行生產數量的預測。達成本發明第二目的所採用之技術手段,係包括單兵式自主移動巡檢載具、中央處理單元、生長環境感測單元及顯示單元。單兵式自主移動巡檢載具於菇類栽培場域的地面通道執行自主巡檢任務。更包括訊號處理模組、行走控制模組及可視化圖表顯示模組。生長環境感測單元隨單兵式自主移動巡檢載具而依序感測各停駐位置的生長環境狀態而產生生長環境感測訊號,由訊號處理模組轉換處理為生長環境數據。可視化圖表顯示模組將生長環境數據轉換為圖表格式資料。生長記錄模組沿著時間軸以依序將各圖表格式資料彙集記錄為生長環境檔案。顯示單元將生長環境檔案顯示為可視化生長環境圖表及生長環境數據。更包含設於該單兵式自主移動巡檢載具上可受訊號處理模組控制而作多軸移動的一移載機構及設於移載機構上的影像擷取單元,使影像擷取單元與生長環境感測單元隨著該單兵式自主移動巡檢載具及移載機構一同載移至每一停 駐位置及每一檢測位置。每一檢測位置設定有一位置編碼。菇類栽培場域並置有複數個菇籃置架,該複數個菇籃置架相互間隔而區隔成該複數個地面通道,每一菇籃置架包括有複數個由下而上分佈的置放位置,每一置放位置供放置一菇籃,每一菇籃置放複數培栽有菇類的太空包。當影像擷取單元抵達停駐位置時,移載機構驅使影像擷取單元依序停留每一檢測位置,使影像擷取單元於每一檢測位置依序拍攝太空包的菇類影像。當生長記錄模組依序將所對應的位置編碼逐一疊加於每一菇類影像上,生長記錄模組沿著時間軸以檔案形式依序將疊加有位置編碼的菇類影像記錄為生長影像檔案。生長狀態辨識模組包含一菇類數量預估模組及內建有複數菇類頭部特徵樣本的特徵資料庫,菇類數量預估模組將菇類影像進行影像處理及裁切邊緣重疊的部分,並對菇類影像做特徵擷取作為菇類頭部特徵,再執行菇類數量計算的影像辨識處理,以將該複數菇類頭部特徵依序輸入至該特徵資料庫,以預測該複數菇類頭部特徵與該菇類頭部特徵樣本的符合機率,當該符合機率大於一預設機率時,則輸出每一檢測位置的菇類數量預估值,再將各該檢測位置的該菇類數量預估值予以累計計算,以得到整個該菇類栽培場域的菇類數量預估資訊。 The second purpose of the present invention is to provide an autonomous mobile artificial intelligence mushroom cultivation monitoring system and method with the function of predicting mushroom production, mainly by using big data collection, artificial intelligence frame selection and training to monitor the mushrooms in the warehouse. Class for production quantity forecasting. The technical means adopted to achieve the second objective of the present invention include a single-soldier autonomous mobile inspection vehicle, a central processing unit, a growth environment sensing unit and a display unit. The single-soldier autonomous mobile inspection vehicle performs autonomous inspection tasks in the ground passage of the mushroom cultivation field. It also includes a signal processing module, a walking control module and a visual chart display module. The growth environment sensing unit sequentially senses the growth environment status of each parking position along with the individual autonomous mobile inspection vehicle to generate growth environment sensing signals, which are converted and processed into growth environment data by the signal processing module. The visual chart display module converts the growth environment data into chart format data. The growth recording module collects and records the data in chart format sequentially along the time axis as a growth environment file. The display unit displays the growth environment file as a visual growth environment chart and growth environment data. It also includes a transfer mechanism and an image capture unit arranged on the transfer mechanism, which can be controlled by the signal processing module to move multi-axis on the individual autonomous mobile inspection vehicle, so that the image capture unit Together with the growth environment sensing unit, it is carried and moved to each stop along with the individual autonomous mobile inspection vehicle and the transfer mechanism. Stationary position and each detection position. Each detection position is set with a position code. A plurality of mushroom basket racks are arranged side by side in the mushroom cultivation field, and the plurality of mushroom basket racks are separated from each other to form the plurality of ground passages, and each mushroom basket rack includes a plurality of bottom-up distribution racks. Placement, each placement position is for placing a mushroom basket, and each mushroom basket is to place a plurality of space bags cultivated with mushrooms. When the image capture unit arrives at the parking position, the transfer mechanism drives the image capture unit to stop at each detection position sequentially, so that the image capture unit sequentially captures images of mushrooms in the space bag at each detection position. When the growth recording module superimposes the corresponding position code on each mushroom image one by one, the growth recording module sequentially records the mushroom images superimposed with the position code in the form of a file along the time axis as a growth image file . The growth status identification module includes a mushroom quantity estimation module and a feature database with multiple mushroom head feature samples built in. The mushroom quantity estimation module performs image processing on mushroom images and crops overlapping edges. part, and extract features from mushroom images as mushroom head features, and then perform image recognition processing for counting the number of mushrooms, so as to sequentially input the multiple mushroom head features into the feature database to predict the number of mushrooms The coincidence probability of the plurality of mushroom head features and the mushroom head feature sample, when the coincidence probability is greater than a preset probability, the estimated value of the number of mushrooms at each detection position is output, and then the number of mushrooms at each detection position The estimated value of the mushroom quantity is calculated cumulatively to obtain the estimated information of the mushroom quantity in the entire mushroom cultivation field.
本發明第三目的,在於提供一種具備熱像儀感測菇類自身生長溫度的自主移動式人工智慧菇類栽培監控系統及方法。達成本發明第四目的所採用之技術手段,係包括單兵式自主移動巡檢載具、中央處理單元、生長環境感測單元及顯示單元。單兵式自主移動巡檢載具於菇類栽培場域的地面通道執行自主巡檢任務。更包含訊號處理模組、行走控制模組及可視化圖表顯示模組,行走路徑設有複數停駐位置。生 長環境感測單元可隨著單兵式自主移動巡檢載具而依序感測各停駐位置的生長環境狀態而產生生長環境感測訊號,由訊號處理模組轉換處理為生長環境數據。可視化圖表顯示模組將生長環境數據轉換為圖表格式資料。生長記錄模組沿著時間軸以依序將各圖表格式資料彙集記錄為生長環境檔案。顯示單元將生長環境檔案顯示為可視化生長環境圖表及生長環境數據。更包含設於該移載機構而位於該影像擷取單元附近的一熱成像儀,當該熱成像儀抵達其中一個該停駐位置時,該訊號處理模組則控制該移載機構驅使該熱成像儀機依序停留該每一該檢測位置,以令該熱成像儀於該每一該檢測位置依序拍攝該太空包的菇類熱影像,當該訊號處理模組依序接收到各該菇類熱影像時,則依序將所對應的該位置編碼逐一疊加於每一該菇類熱影像上,並對每一該菇類熱影像進行色溫分佈的分析處理,以判斷該檢測位置之各該太空包的該菇類溫度是否溫超過預設溫度值,判斷結果為是,則發出與該位置編碼對應的菇類溫度過高的警報訊號。 The third object of the present invention is to provide an autonomous mobile artificial intelligence mushroom cultivation monitoring system and method equipped with a thermal imaging camera to sense the growth temperature of the mushroom itself. The technical means adopted to achieve the fourth objective of the present invention include a single-soldier autonomous mobile inspection vehicle, a central processing unit, a growth environment sensing unit and a display unit. The single-soldier autonomous mobile inspection vehicle performs autonomous inspection tasks in the ground passage of the mushroom cultivation field. It also includes a signal processing module, a walking control module, and a visual chart display module. There are multiple parking positions on the walking path. born The long environment sensing unit can sequentially sense the growth environment status of each parking position along with the individual autonomous mobile inspection vehicle to generate growth environment sensing signals, which are converted and processed into growth environment data by the signal processing module. The visual chart display module converts the growth environment data into chart format data. The growth recording module collects and records the data in chart format sequentially along the time axis as a growth environment file. The display unit displays the growth environment file as a visual growth environment chart and growth environment data. It further includes a thermal imager arranged on the transfer mechanism and located near the image capture unit. When the thermal imager reaches one of the parking positions, the signal processing module controls the transfer mechanism to drive the thermal imager. The imager stays at each of the detection positions in sequence, so that the thermal imager sequentially shoots the mushroom thermal images of the space bag at each of the detection positions. When the signal processing module receives each of the detection positions in sequence In the case of thermal images of mushrooms, the corresponding position codes are sequentially superimposed on each thermal image of mushrooms one by one, and the color temperature distribution of each thermal image of mushrooms is analyzed and processed to determine the location of the detection position. Whether the temperature of the mushrooms in each of the space packs exceeds a preset temperature value, if the judgment result is yes, then an alarm signal corresponding to the position code that the temperature of the mushrooms is too high is sent.
1:菇類栽培場域 1: Mushroom cultivation field
1a:地面通道 1a: Ground access
10:單兵式自主移動巡檢載具 10:Individual-type autonomous mobile inspection vehicle
11:光學雷達 11: Optical radar
12:移載機構 12: transfer mechanism
120:旋轉機構 120: rotating mechanism
121:移載控制模組 121: Transfer control module
121:升降機構 121: lifting mechanism
122:馬達 122: motor
123:滾珠導螺桿 123: Ball lead screw
124:螺帽 124: Nut
125:滑塊 125: slider
126:滑軌 126: slide rail
127:收摺式連桿機構 127: retractable link mechanism
128:驅動座 128:Drive seat
129:移載控制模組 129: Transfer control module
13:影像擷取單元 13: Image capture unit
13:影像擷取單元 13: Image capture unit
14:熱成像儀 14: thermal imager
15:車載電力系統 15: On-board power system
20:中央處理單元 20: Central processing unit
210:菇類數量預估模組 210: Mushroom Quantity Estimation Module
210a:人工智慧深度學習模組 210a: Artificial Intelligence Deep Learning Module
211:特徵資料庫 211: Feature database
212:菇類數量計算演算模型 212: Calculus model for counting the number of mushrooms
22:行走控制模組 22: Walking control module
23:可視化圖表顯示模組 23:Visual chart display module
24:生長記錄模組 24: Growth record module
25:設定模組 25: Setting Module
26:場域圖資建置模組 26: Field map data construction module
27:生長狀態辨識模組 27: Growth state identification module
28:巡檢路徑規劃模組 28: Inspection path planning module
30:生長環境感測單元 30: Growth environment sensing unit
31,32:溫/溼度傳感器 31,32: temperature/humidity sensor
33:二氧化碳濃度傳感器 33: Carbon dioxide concentration sensor
40:顯示單元 40: Display unit
41:可視化生長環境圖表 41: Visualizing Growth Environment Charts
410:環境監視區域 410: Environmental monitoring area
50:菇籃置架 50: Mushroom basket rack
51:菇籃 51: Mushroom Basket
52:太空包 52: Space Pack
60:環境調控設備 60: Environmental control equipment
600:生長環境調控模組 600: Growth environment regulation module
61:控制驅動模組 61:Control drive module
62:無線訊號傳輸模組 62: Wireless signal transmission module
pa:停駐位置 pa: park position
pb:檢測位置 pb: detection position
圖1係本發明具體架構的功能方塊示實施意圖。 Fig. 1 is a schematic implementation diagram of the functional blocks of the specific architecture of the present invention.
圖2係本發明單兵式自主移動巡檢載具於菇類栽培場域巡檢的實施示意圖。 Fig. 2 is a schematic diagram of the implementation of the inspection of the individual soldier autonomous mobile inspection vehicle in the field of mushroom cultivation according to the present invention.
圖3係本發明一種單兵式自主移動巡檢載具的具體實施示意圖。 Fig. 3 is a schematic diagram of a specific implementation of a single-soldier autonomous mobile inspection vehicle of the present invention.
圖4係本發明另一種單兵式自主移動巡檢載具的具體實施示意圖。 Fig. 4 is a specific implementation schematic diagram of another individual soldier autonomous mobile patrol vehicle of the present invention.
圖5係本發明人工智慧深度學習模組於訓練階段的流程實施示意圖。 FIG. 5 is a schematic diagram of the implementation process of the artificial intelligence deep learning module in the training phase of the present invention.
圖6係本發明人工智慧深度學習模組於預測階段的流程實施示意圖。 FIG. 6 is a schematic diagram of the implementation process of the artificial intelligence deep learning module in the prediction stage of the present invention.
圖7係本發明影像擷取單元於菇類栽培場域之檢測位置進行拍攝的實施示意圖。 Fig. 7 is a schematic diagram of the implementation of the image capture unit of the present invention shooting at the detection position of the mushroom cultivation field.
圖8係本發明各監控區域的生長環境圖表及數據的畫面顯示示意圖。 Fig. 8 is a schematic diagram of the screen display of growth environment graphs and data in each monitoring area of the present invention.
圖9係本發明單兵式自主移動巡檢載具的巡檢流程實施示意圖。 Fig. 9 is a schematic diagram of the implementation of the inspection process of the individual autonomous mobile inspection vehicle of the present invention.
圖10本發明菇類栽培場域的熱顯像圖像示意圖。 Fig. 10 is a schematic diagram of a thermal imaging image of a mushroom cultivation field according to the present invention.
圖11係本發明於四個視角點進行菇類數量計算與智慧框選的實施示意圖;圖11a為A點;圖11b為B點;圖11c為C點;圖11d為D點。 Fig. 11 is a schematic diagram of the present invention for calculating the number of mushrooms and intelligent frame selection at four viewpoints; Fig. 11a is point A; Fig. 11b is point B; Fig. 11c is point C; Fig. 11d is point D.
為讓 貴審查委員能進一步瞭解本發明整體的技術特徵與達成本發明目的之技術手段,玆以具體實施例並配合圖式加以詳細說明: In order to allow your review committee to further understand the overall technical characteristics of the present invention and the technical means to achieve the purpose of the present invention, specific embodiments and accompanying drawings are hereby described in detail:
請配合參看圖1~4所示,為達成本發明第一目的之第一實施例,係包括一單兵式自主移動巡檢載具10、一中央處理單元20、一生長環境感測單元30、一顯示單元40、一影像擷取單元13、一場域圖資建置模組26、一生長狀態辨識模組27、一巡檢路徑規劃模組28及一生長環境調控模組600。單兵式自主移動巡檢載具10上設有一行走控制模組22、一移載機構12及一移載控制模組121,該行走控制模組22用以控制該單兵式自主移動巡檢載具10作動,該移載控制模組121用以控制該移載機構12作動。影像擷取單元13用以擷取一菇類栽培場域1內的複數個影像,該複數個影像包括該菇類栽培場域1的至少一地圖影像及在一生長時序中的至少一菇類影像。生長環境感測單元30用以感測該菇類栽培場域內的菇類之數複個生長環境數據,該數複個生長環境數
據包括溫度數據、濕度數據及二氧化碳濃度數據。場域圖資建置模組26用以依據該影像擷取單元13所擷取的該至少一地圖影像而建置成一場域地圖圖資;其中,該場域地圖圖資包括該菇類栽培場域1的至少一地面通道圖資及至少一菇籃置架圖資。生長狀態辨識模組27用以將該至少一菇類影像處理並與一資料庫中預設的相同於該生長時序的基準菇類影像比對;其中,當該至少一菇類影像小於該基準菇類影像且差值大於一第一預定閥值時,判斷該至少一菇類為遲緩生長。巡檢路徑規劃模組28用以依據該至少一地面通道圖資而規劃設定至少一行走路徑及位在該至少一行走路徑上的複數個停駐位置,及依據該至少一菇籃置架圖資而規劃設定至少一檢測路徑及位在該至少一檢測路徑上的複數個檢測位置;該行走控制模組22用以控制該單兵式自主移動巡檢載具10依據該巡檢路徑規劃模組28所規劃的該至少一行走路徑而在至少一預定巡檢期間於該菇類栽培場域1的複數地面通道1a行走及依序於每一該複數個停駐位置pa停駐。其中,該影像擷取單元13及該生長環境感測單元30經由該移載機構12而設於該單兵式自主移動巡檢載具10上,並隨著該單兵式自主移動巡檢載具10依序移動至每一該複數個停駐位置pa;該移載控制模組121用以控制該移載機構12作動以驅使該影像擷取單元13及該生長環境感測單元30依據該巡檢路徑規劃模組28所規劃的該至少一檢測路徑而以各該停駐位置為基點地依序於該菇類栽培場域1的每一該複數個檢測位置pb停駐,進而使該影像擷取單元13得以擷取該至少一菇類影像,及使該生長環境感測單元30得以感測該數複個生長環境數據。生長環境調控模組600用以驅動至少一生長環境調控設備以調控該菇類栽培場域1內的溫度、濕度及二氧化碳濃度;其中,當該生長狀態辨識模組27判斷該至少一菇類為遲緩生長時,該生長環境調控模組600
則依據該資料庫中預設而相符合的一第一生長環境條件參數來調控該菇類栽培場域1內的溫度、濕度及二氧化碳濃度,以加速該至少一菇類生長。中央處理單元20用以整合及處理該行走控制模組22、該移載控制模組121、該影像擷取單元13、該生長環境感測單元30、該場域圖資建置模組26、該生長狀態辨識模組27、該巡檢路徑規劃模組28及該生長環境調控模組600的訊號及資訊。較佳地,當該至少一菇類影像大於該基準菇類影像且差值大於一第二預定閥值時,判斷該至少一菇類為逾速生長;當該生長狀態辨識模組27判斷該至少一菇類為逾速生長時,該生長環境調控模組600則依據該資料庫中預設而相符合的一第二生長環境條件參數來調控該菇類栽培場域1內的溫度、濕度及二氧化碳濃度,以減緩該至少一菇類生長。較佳地,該影像擷取單元13包括至少一LED魚眼補光模組131,該至少一LED魚眼補光模組131用以提供對該影像擷取單元13作動時的補光作用,以提升在暗黑之該菇類栽培場域1內所擷取到的該複數個影像之品質。執行本發明方法,即以該影像擷取單元13擷取一菇類栽培場域1內的複數個影像,該複數個影像包括該菇類栽培場域1的至少一地圖影像及在一生長時序中的至少一菇類影像;以該生長環境感測單元30感測該菇類栽培場域1內的菇類之數複個生長環境數據,該數複個生長環境數據包括溫度數據、濕度數據及二氧化碳濃度數據;以該場域圖資建置模組26依據該影像擷取單元13所擷取的該至少一地圖影像而建置成一場域地圖圖資,該場域地圖圖資包括該菇類栽培場域1的至少一地面通道圖資及至少一菇籃置架圖資;以該生長狀態辨識模組27將該至少一菇類影像處理並與一資料庫中預設的相同於該生長時序的基準菇類影像比對;其中,當該至少一菇類影像小於該基準菇類影像且差值大於一第一預定閥值時,判斷該至少一菇類為遲緩生
長;以該巡檢路徑規劃模組28依據該至少一地面通道圖資而規劃設定至少一行走路徑及位在該至少一行走路徑上的複數個停駐位置pa,及依據該至少一菇籃置架圖資而規劃設定至少一檢測路徑及位在該至少一檢測路徑上的複數個檢測位置pb;該行走控制模組22用以控制該單兵式自主移動巡檢載具10依據該巡檢路徑規劃模組28所規劃的該至少一行走路徑而在至少一預定巡檢期間於該菇類栽培場域的複數地面通道1a行走及依序於每一該複數個停駐位置pa停駐;使該影像擷取單元13及該生長環境感測單元30經由該移載機構12而設於該單兵式自主移動巡檢載具10上,並隨著該單兵式自主移動巡檢載具10依序移動至每一該複數個停駐位置pa;該移載控制模組121用以控制該移載機構12作動以驅使該影像擷取單元13及該生長環境感測單元30依據該巡檢路徑規劃模組28所規劃的該至少一檢測路徑而以各該停駐位置為基點地依序於該菇類栽培場域1的每一該複數個檢測位置pb停駐,進而使該影像擷取單元13得以擷取該至少一菇類影像,及使該生長環境感測單元30得以感測該數複個生長環境數據;以該生長環境調控模組600驅動至少一生長環境調控設備60以調控該菇類栽培場域1內的溫度、濕度及二氧化碳濃度;其中,當該生長狀態辨識模組27判斷該至少一菇類為遲緩生長時,該生長環境調控模組600則依據該資料庫中預設而相符合的一第一生長環境條件參數來調控該菇類栽培場域1內的溫度、濕度及二氧化碳濃度,以加速該至少一菇類生長;其中,當該生長狀態辨識模組27判斷該至少一菇類為逾速生長時,則依據該資料庫中預設而相符合的一第二生長環境條件參數來調控該菇類栽培場域1內的溫度、濕度及二氧化碳濃度,以減緩該至少一菇類生長;及以該中央處理單元20整合及處理該行走控制模組22、該移載控制模組121、該影像擷取單元13、該生長環境
感測單元30、該場域圖資建置模組26、該生長狀態辨識模組27、該巡檢路徑規劃模組28及該生長環境調控模組600的訊號及資訊。
Please refer to Figures 1 to 4, in order to achieve the first purpose of the present invention, the first embodiment includes a single-soldier autonomous
請配合參看圖1~4所示,一種具體實施例中,該單兵式自主移動巡檢載具10係於每一預設間隔時間而於一菇類栽培場域1的複數地面通道1a執行自主巡檢任務。內建有一行走路徑的一訊號處理模組21,更包括有一可視化圖表顯示模組23、一生長記錄模組24及一無線訊號傳輸模組62,該中央處理單元20整合及處理該可視化圖表顯示模組、該生長記錄模組24及該無線訊號傳輸模組62的訊號及資訊。行走路徑設定有複數停駐位置pa。行走控制模組22受訊號處理模組21的控制而驅使單兵式自主移動巡檢載具10沿著行走路徑而於複數地面通道1a行走及依序於各停駐位置pa停駐。該生長環境感測單元30經由該移載機構12而設於單兵式自主移動巡檢載具10上,可隨著單兵式自主移動巡檢載具10執行巡檢任務而依序感測各停駐位置pa所對應的各檢測位置pb之生長環境狀態而產生生長環境感測訊號,並由訊號處理模組21轉換處理生長環境感測訊號相應的生長環境數據,該生長環境數據係選自溫度、濕度以及二氧化碳濃度的其中至少二種。可視化圖表顯示模組23用以將生長環境數據轉換為可視化圖表方式來顯示的圖表格式資料。該生長記錄模組24設於單兵式自主移動巡檢載具10,並沿著一時間軸以每一預設間隔時間為一個檔案的方式依序將各圖表格式資料及各生長環境數據彙集記錄為生長環境檔案。該可視化圖表顯示模組23將該可視化生長環境圖表及該生長環境數據經由該無線訊號傳輸模組62傳輸至一電子裝置,以供該電子裝置的一顯示單元加以顯示。
Please refer to Figures 1 to 4. In a specific embodiment, the individual soldier-type autonomous
請配合參看圖8所示的實施例,本實施例為上述第一實施例更為具體描述可視化生長環境圖表41技術的具體實施例,該可視
化生長環境圖表41等分地劃分出呈陣列且逐一對應於該複數個停駐位置的複數環境監視區域410,每一該複數個環境監視區域410分別標示有所對應之每一該複數個停駐位置的編碼(pa1,pa2,pa3,pa4)。每一該複數個環境監視區域410劃分出呈陣列且逐一對應於該複數個檢測位置的複數個數據監視區域,每一該複數個數據監視區域標示有所對應之每一該複數個檢測位置的編碼(pb101,pb102,pb103,pb104,pb201,pb202,pb203,pb204,pb301,pb302,pb303,pb304,pb401,pb402,pb403,pb404)及顯示有所對應的每一該複數個檢測位置之即時的溫度數據、濕度數據及二氧化碳濃度數據。
Please refer to the embodiment shown in Figure 8. This embodiment is a specific embodiment of the above-mentioned first embodiment to more specifically describe the visualization of the
請配合參看圖1、3及圖4所示的實施例,本實施例為上述第一實施例更為具體描述生長環境檔案與路徑規劃等技術的具體實施例,該生長環境檔案係以一天為一個單位而按照日期前後來排列,該中央處理單元依據所需而讀取其中一個生長環境檔案時,則於每一環境監視區域410各自顯示有沿著時間軸而變化的溫度曲線圖、濕度曲線圖以及二氧化碳濃度曲線圖;該單兵式自主移動巡檢載具10設有至少一光學雷達11,該訊號處理模組21透過至少一光學雷達11,並搭配ROS SLAM演算法進行場域圖資的建置,再將建立的地圖導入行走路徑中,以於菇類栽培場域1執行自主巡檢任務,並於複數地面通道1a記錄下作為複數停駐位置pa的路徑點。
Please cooperate with referring to the embodiment shown in Fig. 1, 3 and Fig. 4, this embodiment is the concrete embodiment that more specifically describes technologies such as growth environment file and path planning in the above-mentioned first embodiment, and this growth environment file system is based on one day A unit is arranged according to the date, and when the central processing unit reads one of the growth environment files according to the needs, each
請配合參看圖1~4及圖7~8所示,為達成本發明第二目的之第二實施例,本實施例除了包括上述第一實施例的整體技術內容之外,更包含設於單兵式自主移動巡檢載具10上可受訊號處理模組21控制而作多軸移動的移載機構12及設於移載機構12上的影像擷取單元13,使影像擷取單元13與生長環境感測單元30隨著單兵式自主移動巡檢載具10及移載機構12一同載移至每一停駐位置pa;該訊號處理模組
21將每一停駐位置pa設定有複數檢測位置pb,每一檢測位置pb設定有一位置編碼。該菇類栽培場域1並置有複數個菇籃置架50,該複數個菇籃置架50相互間隔而區隔成複數個地面通道1a,每一該複數個菇籃置架50包括有複數個由下而上分佈的置放位置,每一置放位置供放置一菇籃51,每一菇籃51置放複數培栽有菇類的太空包52;當影像擷取單元13抵達其中一個停駐位置pa時,該訊號處理模組21則控制移載機構12驅使影像擷取單元13依序停留每一檢測位置pb,以令影像擷取單元13於每一檢測位置pb依序拍攝太空包52的菇類影像,當訊號處理模組21依序接收到各菇類影像時,則依序將所對應的位置編碼逐一疊加於每一菇類影像上,該訊號處理模組21驅使生長記錄模組沿著時間軸以檔案形式依序將疊加有位置編碼的各菇類影像記錄為生長影像檔案。該訊號處理模組21包含一菇類數量預估模組210及一內建有複數菇類頭部特徵樣本的特徵資料庫211,當菇類數量預估模組210接收到影像擷取單元13於一個檢測位置pb所拍攝的菇類影像時,則進行影像處理及裁切邊緣重疊的部分,並對菇類影像依序做特徵擷取為複數菇類頭部特徵,再執行菇類數量計算的影像辨識處理,以將複數菇類頭部特徵依序輸入至特徵資料庫211,以預測複數菇類頭部特徵與菇類頭部特徵樣本的符合機率,當符合機率大於一預設機率時,則輸出檢測位置pb的菇類數量預估值;當菇類數量預估模組210接收到影像擷取單元13於下一個檢測位置pb所拍攝的菇類影像時,則重覆上述的影像處理、裁切及影像辨識處理等步驟,並計算輸出下一個檢測位置pb的菇類數量預估值,直到所有檢測位置pb的菇類影像皆完成上述的影像處理、裁切及影像辨識處理等步驟,再將各檢測位置pb的菇類數量預估值予以累計計算,於是即可得到整個菇類栽培場域1的菇類數量預估資訊。再請配合參看圖7所示,IA即影像擷取單元13於其中一個檢測位置pb的拍攝範圍,亦即菇類影像的拍攝範圍。
該移載機構12為一多軸移載機構;該行走控制模組22包括有一避障模組,該避障模組包括有至少一光學雷達,該至少一光學雷達於該單兵式自主移動巡檢載具10行走時感測該至少一行走路徑上是否有障礙物,當感測到在抵達下一個該複數個停駐位置之前有一障礙物時則修正該至少一行走路徑,使該該單兵式自主移動巡檢載具10能抵達該下一個該複數個停駐位置,再繼續該至少一行走路徑的後續行程;其中,當該障礙物位在一該複數個停駐位置時,該行走控制模組22控制該單兵式自主移動巡檢載具10於鄰近位有該障礙物的一該複數個停駐位置之一臨停位置處停駐,該移載控制模組121依據該臨停位置的座標與位有該障礙物的一該複數個停駐位置之座標比對而獲得一補償位移座標參數,並依據該補償位移座標參數來控制該多軸移載機構作動而將該該影像擷取單元13及該生長環境感測單元30依序地移載至以位有該障礙物的一該複數個停駐位置為基點的每一該複數個檢測位置。
Please refer to Figures 1-4 and Figures 7-8, in order to achieve the second embodiment of the second purpose of the present invention, this embodiment not only includes the overall technical content of the above-mentioned first embodiment, but also includes The military-style autonomous
請參看圖5~6所示,本實施例為上述第二實施例採用人工智慧影像辨識技術的具體實施例。該生長狀態辨識模組27包括一菇類數量預估模組210,該菇類數量預估模組210包括一人工智慧深度學習模組210a,該人工智慧深度學習模組210a依據一訓練學習步驟而於該特徵資料庫211建立一菇類數量計算演算模型212,於菇類數量計算演算模型212型輸入巨量的菇類頭部特徵樣本、菇類特徵參數、人工智慧框選參數及影像辨識參數,並由菇類數量計算演算模型212測試各菇類影像的影像辨識正確率,再判斷各菇類影像的影像辨識正確率是否足夠,當判斷結果為是,則將辨識結果輸出及儲存;當判斷結果為否,則使該菇類數量計算演算模型212自我修正學習;該人工智慧深度學習模組210a執行影像辨識處理時,則執行一預測階段步驟,係於菇類數量計算演算模型212依序輸入即時連續輸入已經裁切的菇類影像,並由菇類數量計算
演算模型212預測辨識出所即時輸入之菇類影像所代表的菇類數量預估值,然後再預測辨識出整個菇類栽培場域1的菇類數量預估資訊。
Please refer to FIGS. 5-6 , this embodiment is a specific embodiment of the second embodiment using artificial intelligence image recognition technology. The growth
請配合參看圖4所示,為達成本發明第三目的之第三實施例,本實施例除了包括上述第一、第二實施例的整體技術內容之外,該移載機構12包含可做水平向旋轉的一旋轉機構120及可做縱向線性伸縮位移而位於旋轉機構120上的一升降機構121,該升降機構121包括一馬達122、一具有二互為反向之螺紋段的滾珠導螺桿123、二螺帽124、一滑軌126、一可於滑軌126滑移的滑塊125、一收摺式連桿機構127及一設於滑塊125上的驅動座128;該驅動座128固設影像擷取單元13,該二螺帽124分別套設於滾珠導螺桿123互為反向之的螺紋段上,以驅使二螺帽124沿著一位於滾珠導螺桿123二側的反向螺紋段做反向位移。該收摺式連桿機構127之二動力輸入端與二螺帽124可轉動地樞接,其動力輸出端則與驅動座128樞接;當馬達122驅動滾珠導螺桿123轉動而驅使二螺帽124往外側做反向位移時,則連動收摺式連桿機構127呈收摺狀態,並連動驅動座128及滑塊125沿著滑軌126往下方移動(即收縮動作);當馬達122驅動滾珠導螺桿123反向轉動而驅使二螺帽124往內側做反向位移時,則連動收摺式連桿機構127呈伸展狀態,並連動驅動座128及滑塊125沿著滑軌126往上方移動(即伸展動作)。
Please refer to the third embodiment shown in Figure 4, in order to achieve the third embodiment of the third purpose of the present invention, in addition to the overall technical content of the above-mentioned first and second embodiments, the
請配合參看圖1、3及圖4所示,為達成本發明第四目的之第四實施例,本實施例除了包括上述第一、第二實施例的整體技術內容之外,更包含設於移載機構12而位於影像擷取單元13附近的一熱成像儀14,當該熱成像儀14抵達其中一個該停駐位置pa時,該訊號處理模組21則控制該移載機構12驅使熱成像儀14依序停留每一檢測位置pb,以令熱成像儀14於每一檢測位置pb依序拍攝太空包52的菇類熱影像,當訊號處理模組21依序接收到各菇類熱影像時,則依序將所對應
的位置編碼逐一疊加於每一菇類熱影像上,並對每一菇類熱影像進行色溫分佈的分析處理,以判斷檢測位置pb之各太空包52的菇類溫度是否溫超過預設溫度值,判斷結果為是,則發出與位置編碼對應的菇類溫度過高的警報訊號。
Please refer to Fig. 1, 3 and Fig. 4, in order to achieve the fourth embodiment of the fourth purpose of the present invention, in addition to the overall technical content of the above-mentioned first and second embodiments, this embodiment also includes The
請配合參看圖1所示,為達成本發明第五目的之第五實施例,本實施例除了包括上述第一實施例的整體技術內容之外,更包含一用以調控菇類栽培場域1之生長環境狀態的生長環境調控設備60及一控制驅動模組61。該生長環境調控模組600包括可供選擇設定為一加速生長模式、一正常生長模式及一減緩生長模式的一設定模組25。當以設定模組25選擇為該加速生長模式時,該生長環境調控模組600則產生一加速生長的控制指令,並透過一無線訊號傳輸模組62(如藍芽通訊模組;但不以此為限)傳輸給控制驅動模組61,以控制驅動模組61來驅動生長環境調控設備60,使菇類栽培場域1的生長環境條件符合加速生長模式。當以設定模組25選擇為正常生長模式時,該生長環境調控模組600則產生一正常生長的控制指令,並透過無線訊號傳輸模組62傳輸給控制驅動模組61,以控制驅動模組61來驅動環境調控設備60,使菇類栽培場域1的生長環境條件符合正常生長模式。當以設定模組25選擇減緩生長模式時,該生長環境調控模組600則產生減緩生長的控制指令,並透過無線訊號傳輸模組62傳輸給該控制驅動模組61,以該控制驅動模組61來驅動該生長環境調控設備60,使菇類栽培場域1的生長環境條件符合減緩生長模式。
Please refer to the fifth embodiment shown in Fig. 1, in order to achieve the fifth embodiment of the present invention, this embodiment not only includes the overall technical content of the above-mentioned first embodiment, but also includes a
再者,基於降低巡檢人力的需求,係利用現有菇房場域配置設計開發一台具有自我導航、避障、多點巡航及各式感測器資料蒐集並上傳資料庫的自主移動式智慧物聯模組,該模組是利用影像辨識與人工智慧技術,來達到菇類生產環境監控與生產數量預測的目的。本發明之自主移動式智慧物聯模組是基於移動機器人底盤架構來進行開發設計,本模組是 基於Ubuntu建置機器人操作系統(Robot Operating System,ROS),透過光學雷達(Lidar)來知道與障礙物的位置與距離,再與模組上的慣性測量單元(Inertial MeasurementUnit,IMU)、直流減速馬達帶霍爾編碼器上的霍爾回授的參數結合判斷,以達到自我導航、避障、多點巡航的目的。生產預測數量係利用大數據收集、人工智慧框選及訓練,讓自主移動式智慧物聯模組在生產庫房間拍照,即可對於庫房內的杏鮑菇外型進行生產數量預測的目的。 Furthermore, based on the need to reduce manpower for inspections, an autonomous mobile smart device with self-navigation, obstacle avoidance, multi-point cruise, and various sensor data collection and uploading databases is designed and developed by using the existing mushroom house field configuration. IoT module, this module uses image recognition and artificial intelligence technology to achieve the purpose of mushroom production environment monitoring and production quantity prediction. The autonomous mobile intelligent IoT module of the present invention is developed and designed based on the mobile robot chassis architecture. This module is Build a robot operating system (Robot Operating System, ROS) based on Ubuntu, know the position and distance to obstacles through Lidar, and then communicate with the inertial measurement unit (Inertial Measurement Unit, IMU) and DC geared motor on the module Combined with the parameters of the Hall feedback on the Hall encoder to achieve the purpose of self-navigation, obstacle avoidance, and multi-point cruise. The production forecast quantity is based on big data collection, artificial intelligence frame selection and training, so that the autonomous mobile intelligent IoT module can take pictures in the production warehouse room, so as to predict the production quantity of the Pleurotus eryngii in the warehouse.
本發明設計發展的自主移動式智慧物聯模組動作流程控制實施如圖9所示,將模組上的小型電腦(即訊號處理模組)啟動後,利用底盤驅動系統與光學雷達並搭配ROS SLAM演算法進行場域圖資的建置,並把建立的地圖導入ROS Navigation進行場域自主導航並設置記錄路徑點,透過ROS Python依照建立的地圖與記錄的路徑點並搭配影像擷取系統中的影像擷取單元13(即網路攝影機;Logitech C525)與移載機構12(即手臂驅動系統)來進行拍照以實現場域多點巡航與拍照的需求,再把透過模組拍攝的照片導入物件偵測系統以進行杏鮑菇的物件偵測與數量的計算。另外,模組開啟後,環境感測系統也隨之啟動,當模組抵達設置的路徑點時將觸發環境感測系統,利用模組上裝載的工業級感測器感測場域環境中的溫度、濕度及二氧化碳濃度,再透過ESP32單晶片電腦將感測器的數值傳送至自行建置的資料庫伺服器(MySQL)與生產管理者的LINE Notify,再利用可視化軟體(Grafana)將數值以圖表的方式顯示,以達到環境感測資料收集、儲存與顯示的目的。 The implementation of the action flow control of the autonomous mobile intelligent IoT module designed and developed by the present invention is shown in Figure 9. After the small computer (i.e., the signal processing module) on the module is started, the chassis drive system and optical radar are used together with ROS The SLAM algorithm constructs the map data of the field, and imports the established map into ROS Navigation for autonomous navigation of the field and sets the recorded path points. Through ROS Python, the established map and the recorded path points are matched with the image capture system The image capture unit 13 (that is, the network camera; Logitech C525) and the transfer mechanism 12 (that is, the arm drive system) are used to take pictures to realize the needs of multi-point cruising and taking pictures in the field, and then import the pictures taken through the module into the The object detection system is used for object detection and quantity calculation of Pleurotus eryngii. In addition, after the module is turned on, the environmental sensing system is also activated. When the module reaches the set waypoint, the environmental sensing system will be triggered, and the industrial-grade sensor mounted on the module will be used to sense the environment in the field environment. Temperature, humidity and carbon dioxide concentration, and then send the sensor value to the self-built database server (MySQL) and LINE Notify of the production manager through the ESP32 single-chip computer, and then use the visualization software (Grafana) to transfer the value to Displayed in the form of charts to achieve the purpose of collecting, storing and displaying environmental sensing data.
本發明硬體系統架構如圖1所示,並可區分為訊號處理模組21(即上層系統單元)與移載機構12(即下層執行單元)。上層系統單元採用小型電腦(ASUS Mini PC PB60G)作為整個系統架構的核心,其小型電腦運行Ubuntu作業系統以及ROS機器人操作系統,主要任務有場域SLAM地圖建置、場域圖資導航、電腦視覺用於物件判斷,並透過ROS與USB通訊對於 下層執行單元發送命令、接收並處理下層單元的數據。下層執行單元分為底盤驅動系統、環境感測系統、手臂驅動系統。底盤驅動系統透過廠商開發的STM32開發板,主要負責直流馬達的控制、Encoder訊號的處理、MPU9250九軸加速度傳感器訊號處理。環境感測系統透過裝載在智慧物聯模組上的溫溼度傳感器31,32(eYc THS13)與二氧化碳濃度傳感器33(eYc GS43),再利用ESP32 DOIT DEVKIT單晶片電腦把傳感器的數值傳送至自行建置的資料庫與生產管理者的LINE Notify,以達到庫房的環境監控。手臂驅動系統採用Open MANIPULATORX手臂搭配U2D2控制板可以透過USB來讓主電腦透過命令方式控制手臂的姿態。 The hardware system architecture of the present invention is shown in FIG. 1 , and can be divided into a signal processing module 21 (ie the upper system unit) and a transfer mechanism 12 (ie the lower execution unit). The upper system unit uses a small computer (ASUS Mini PC PB60G) as the core of the entire system architecture. The small computer runs the Ubuntu operating system and the ROS robot operating system. The main tasks include field SLAM map construction, field map data navigation, and computer vision. It is used for object judgment, and communicates with USB through ROS for The lower-level execution unit sends commands, receives and processes data from the lower-level units. The lower execution unit is divided into chassis drive system, environment sensing system, and arm drive system. The chassis drive system is mainly responsible for the control of the DC motor, the processing of the Encoder signal, and the signal processing of the MPU9250 nine-axis acceleration sensor through the STM32 development board developed by the manufacturer. The environmental sensing system uses the temperature and humidity sensors 31, 32 (eYc THS13) and carbon dioxide concentration sensors 33 (eYc GS43) loaded on the smart IoT module, and then uses the ESP32 DOIT DEVKIT single-chip computer to transmit the sensor values to the self-built The configured database and LINE Notify of the production manager to achieve environmental monitoring of the warehouse. The arm drive system uses the Open MANIPULATORX arm with the U2D2 control board to allow the host computer to control the arm's posture through commands through USB.
如圖3所示,車載電力系統15分為兩部分,底盤電源與系統負載電源。底盤電源負責供應底盤控制板與馬達的電源,電池採用3顆18650鋰電池串接而成,電壓為12.6V。系統負載電源負責供應車載電腦、環境感測器、手臂的電源,採用18650鋰電池以6串2並的方式,電壓為25.2V,再透過降壓模組降為20V供應給車載電腦使用與12V供應給感測器與手臂系統使用。
As shown in FIG. 3 , the on-
本自主移動式智慧物聯模組為了適應菇房黑暗無光的環境,在模組上裝載了一顆2.5W的魚眼LED,透過Python控制繼電器開啟與關閉LED。當模組進行場域多點巡航與拍照時,LED會自行開啟,巡航結束會自行關閉。 In order to adapt to the dark environment of the mushroom house, this autonomous mobile smart IoT module is equipped with a 2.5W fisheye LED on the module, and the relay is controlled to turn on and off the LED through Python. When the module is performing multi-point cruise and taking pictures in the field, the LED will turn on automatically, and it will turn off automatically when the cruise is over.
本發明為了計算太空包杏鮑菇的數量,係基於YOLO V4演算法加上計數功能,完成圖像物件偵測並數量計數功能,如圖11所示。以圖說明本發明所設計發展的自主移動式智慧物聯模組的硬體配置狀況與重要尺寸圖。 In order to calculate the number of Pleurotus eryngii in the space pack, the present invention is based on the YOLO V4 algorithm plus a counting function to complete the image object detection and quantity counting function, as shown in Figure 11. The hardware configuration and important dimensions of the autonomous mobile smart IoT module designed and developed by the present invention are illustrated with pictures.
本發明利用裝載在自主移動式智慧物聯模組上的熱成像儀14,在農業試驗所場域與模擬測試場域進行杏鮑菇太空包的溫度量測,能
夠記錄菇體的溫度及生長狀況,提供給爾後進行菇類生長預測的數據提供。在農業試驗所擷取的熱顯像圖像中,菇體的溫度落在16℃~16.5℃間,從文獻得知,在此溫度區間的子實體正常生長發育,如圖10所示。
The present invention utilizes the
本發明利用先前在場域進行場域多點巡航與拍照所攝取到的照片,透過自行訓練的杏鮑菇深度學習模型,進行杏鮑菇照片的多點巡航與拍照目標檢測與數量的計算,數量計算的結果於照片的左上角,如表4.3所示。本發明將照片給予8位人員利用肉眼計數照片中杏鮑菇的數量並與自行訓練的神經網路模型進行比對誤差及計算準確率,其結果如圖11所示,四張照片中的最高眾數誤差為3,最低準確率為82.4%。 The present invention utilizes the photographs taken during multi-point cruising and photographing in the field, and through the self-trained Pleurotus eryngii deep learning model, the multi-point cruising and photographing target detection and quantity calculation of Pleurotus eryngii photos are carried out. The result of the quantity calculation is shown in the upper left corner of the photo, as shown in Table 4.3. The present invention gives photos to 8 persons to count the number of Pleurotus eryngii in the photos with the naked eye and compares the error and calculation accuracy with the self-trained neural network model. The results are shown in Figure 11, the highest among the four photos The mode error is 3, and the lowest accuracy rate is 82.4%.
表一
因此,藉由上述具體實施例的詳細說明,本發明確實具備下列所述的特點: Therefore, by the detailed description of the above specific embodiments, the present invention does possess the following characteristics:
1.本發明確實可以藉由可視化生長環境圖表的機能設置,以供監控者以更為快速地解讀出菇類生長環境資訊,以營建出更為適合菇類栽種生長的環境。 1. The present invention can indeed use the function setting of the visualized growth environment chart for the monitor to interpret the information of the mushroom growth environment more quickly, so as to create a more suitable environment for mushroom planting and growth.
2.本發明確實具備菇類產量預測功能的菇類栽培物聯網監測系統及方法,主要是可以利用大數據收集、人工智慧框選及訓練, 以對庫房內的菇類進行生產數量的預測。 2. The present invention does have the mushroom cultivation Internet of Things monitoring system and method with the function of mushroom yield prediction, mainly by using big data collection, artificial intelligence frame selection and training, To predict the production quantity of mushrooms in the warehouse.
3.本發明確實可以讓影像擷取單元可以穩定地增加縱向伸縮位移距離的自菇類栽培物聯網監測系統及方法。 3. The present invention can indeed allow the image capture unit to stably increase the longitudinal telescopic displacement distance of the self-mushroom cultivation Internet of Things monitoring system and method.
4.本發明確實具備具備熱像儀感測菇類自身生長溫度的功能。 4. The present invention does have the function of thermal imager sensing the growth temperature of the mushroom itself.
5.本發明具備可以因應市場需求而加快或減緩菇類生長速度的功能。 5. The present invention has the function of accelerating or slowing down the growth rate of mushrooms in response to market demands.
以上所述,僅為本發明之可行實施例,並非用以限定本發明之專利範圍,凡舉依據下列請求項所述之內容、特徵以及其精神而為之其他變化的等效實施,皆應包含於本發明之專利範圍內。本發明所具體界定於請求項之結構特徵,未見於同類物品,且具實用性與進步性,已符合發明專利要件,爰依法具文提出申請,謹請 鈞局依法核予專利,以維護本申請人合法之權益。 The above is only a feasible embodiment of the present invention, and is not intended to limit the patent scope of the present invention. Any equivalent implementation of other changes based on the content, characteristics and spirit of the following claims should be Included in the patent scope of the present invention. The structural features of the invention specifically defined in the claims are not found in similar items, and are practical and progressive, and have met the requirements of an invention patent. I file an application in accordance with the law. I would like to ask the Jun Bureau to approve the patent in accordance with the law to maintain this invention. The legitimate rights and interests of the applicant.
11:光學雷達 11: Optical radar
12:移載機構 12: transfer mechanism
129:移載控制模組 129: Transfer control module
13:影像擷取單元 13: Image capture unit
14:熱成像儀 14: thermal imager
20:中央處理單元 20: Central processing unit
210:菇類數量預估模組 210: Mushroom Quantity Estimation Module
211:特徵資料庫 211: Feature database
22:行走控制模組 22: Walking control module
23:可視化圖表顯示模組 23:Visual chart display module
24:生長記錄模組 24: Growth record module
25:設定模組 25: Setting Module
26:場域圖資建置模組 26: Field map data construction module
27:生長狀態辨識模組 27: Growth state identification module
28:巡檢路徑規劃模組 28: Inspection path planning module
30:生長環境感測單元 30: Growth environment sensing unit
31,32:溫/溼度傳感器 31,32: temperature/humidity sensor
33:二氧化碳濃度傳感器 33: Carbon dioxide concentration sensor
40:顯示單元 40: Display unit
60:環境調控設備 60: Environmental control equipment
600:生長環境調控模組 600: Growth environment regulation module
61:控制驅動模組 61:Control drive module
62:無線訊號傳輸模組 62: Wireless signal transmission module
Claims (9)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW111128426A TWI806721B (en) | 2022-07-28 | 2022-07-28 | Autonomous Mobile Artificial Intelligence Mushroom Cultivation Monitoring System and Method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW111128426A TWI806721B (en) | 2022-07-28 | 2022-07-28 | Autonomous Mobile Artificial Intelligence Mushroom Cultivation Monitoring System and Method |
Publications (2)
Publication Number | Publication Date |
---|---|
TWI806721B true TWI806721B (en) | 2023-06-21 |
TW202405827A TW202405827A (en) | 2024-02-01 |
Family
ID=87803229
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
TW111128426A TWI806721B (en) | 2022-07-28 | 2022-07-28 | Autonomous Mobile Artificial Intelligence Mushroom Cultivation Monitoring System and Method |
Country Status (1)
Country | Link |
---|---|
TW (1) | TWI806721B (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112230696A (en) * | 2020-10-14 | 2021-01-15 | 王峰 | Fruiting workshop intelligent temperature and humidity control and uniform distribution equipment and using method thereof |
TWI719924B (en) * | 2020-07-27 | 2021-02-21 | 國立虎尾科技大學 | Mushroom growth monitoring system and method |
CN113575285A (en) * | 2021-08-02 | 2021-11-02 | 嘉沁农业科技有限公司 | 5G data interconnection crab flavor mushroom automated production system |
CN114515663A (en) * | 2022-01-11 | 2022-05-20 | 北京市农林科学院信息技术研究中心 | Mushroom spraying control method, device and system |
-
2022
- 2022-07-28 TW TW111128426A patent/TWI806721B/en active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TWI719924B (en) * | 2020-07-27 | 2021-02-21 | 國立虎尾科技大學 | Mushroom growth monitoring system and method |
CN112230696A (en) * | 2020-10-14 | 2021-01-15 | 王峰 | Fruiting workshop intelligent temperature and humidity control and uniform distribution equipment and using method thereof |
CN113575285A (en) * | 2021-08-02 | 2021-11-02 | 嘉沁农业科技有限公司 | 5G data interconnection crab flavor mushroom automated production system |
CN114515663A (en) * | 2022-01-11 | 2022-05-20 | 北京市农林科学院信息技术研究中心 | Mushroom spraying control method, device and system |
Also Published As
Publication number | Publication date |
---|---|
TW202405827A (en) | 2024-02-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102914967B (en) | Autonomous navigation and man-machine coordination picking operating system of picking robot | |
CN108189043B (en) | Inspection method and inspection robot system applied to high-speed rail machine room | |
CN207139822U (en) | Data center's crusing robot | |
CN112414457B (en) | Automatic intelligent inspection method based on transformer substation work | |
Khan et al. | GPS guided autonomous navigation of a small agricultural robot with automated fertilizing system | |
US11406061B2 (en) | Automated walnut picking and collecting method based on multi-sensor fusion technology | |
CN109460029A (en) | Livestock and poultry cultivation place inspection mobile platform and its control method | |
CN109900280A (en) | A kind of livestock and poultry information Perception robot and map constructing method based on independent navigation | |
CN109699300A (en) | A kind of control system and its control method of intelligence citrus picking machine | |
CN108803613A (en) | A kind of intelligent crusing robot control system | |
Silwal et al. | Bumblebee: A Path Towards Fully Autonomous Robotic Vine Pruning. | |
CN105867112B (en) | A kind of intelligent vehicle and its control method of the control algolithm based on parameter automatic optimization | |
CN117111660A (en) | Unattended intelligent granary system and method | |
CN116117807A (en) | Chilli picking robot and control method | |
TWI806721B (en) | Autonomous Mobile Artificial Intelligence Mushroom Cultivation Monitoring System and Method | |
TWI793051B (en) | Monitoring System and Method of Internet of Things for Mushroom Cultivation | |
CN117546681A (en) | Collaborative work method, system and platform for guiding robot to pick and transport fruits | |
CN219161992U (en) | Intelligent plant monitoring robot | |
CN115890677B (en) | Dead chicken picking robot for standardized cage chicken house and method thereof | |
CN114931112B (en) | Sow body ruler detection system based on intelligent inspection robot | |
CN115589845A (en) | Intelligent cotton picking robot and cotton picking operation path planning method thereof | |
CN116466714A (en) | Control system and method for AGV (automatic guided vehicle) | |
KR102264200B1 (en) | Crop growth information monitoring system | |
CN115380711A (en) | Automatic picking robot, automatic picking method, automatic picking apparatus, and storage medium | |
CN111367289A (en) | Artificial intelligence equipment |