Vinnerliste

Når en har en meget komplisert plen på 750 m2 og robotklipper uten sporing så dukker der jo fort opp behov for litt snekring på gutterommet... Har hatt meg en del turer rundt i hagen for å lete etter klipperen som har satt seg fast en eller annen plass så litt hjelp til å lete og, ikke minst, se hvilke områder som klippes er ganske greit... Nå har jeg jobbet noe tid med dette og det er på tide å dokumentere litt av prosessen i tilfelle andre vil gjøre noe lignende... 1: Skaffe kart over hagen. Her har jeg brukt kommunekart.com og "Tegn i kart" for å lage et omriss av plenen. Etter omrisset er laget lagres det til fil i kml format. For enkelhets skyld konverterer jeg denne til .geojson format i online konverteringsprogram. Pr i dag ser omrisset slik ut i kommunekart.com: .geojson filen importeres manuelt i Node-RED. Dette er jo en mer eller mindre engangs sak... 2: Innhente GPS-data og sende til MQTT. ESP32 med NEO-6 GPS koster "next to nothing" og er en grei løsning for dette bruket. Har slitt en del med Arduino-IDE så derfor havnet jeg på MicroPython denne gang. Siden jeg har 2 separate prosesser gående samtidig har jeg valgt å bruke uasyncio for multitasking. Task#1: Innhente data fra GPS og lagre posisjonsdata i fil. Task#2: På gitte intervall, pt hvert 30 sek, koble på wlan og sende data fra datafilen til MQTT broker. I hagen min er der noen wlan dødsoner og derfor nyttig å gjøre opplasting litt i rykk og napp når wlan er tilgjengelig. 3: Behandle GPS-data i Node-RED. Etterhvert som data mottas fra MQTT lagres de i et array of objects i flow-variabel og deretter tegnes det opp i en chart node. Array med data begrenses til et passende antall posisjoner, pt. 3000. Eldre data slettes automatisk. Prototype GPS-tracker: Første sporing: Node-RED: [{"id":"7b7be9e90c83eaff","type":"group","z":"c0e718067b85a8cc","name":"Lagre GPS-data","style":{"label":true},"nodes":["43d583213bfaa02f","2218191fe090798b","55b21eb55f283b61","8c73af56d7fdc323","36f55a2e222b2326","c466d510d0476f0e","93ac548a0bcda12d"],"x":24,"y":779,"w":832,"h":192},{"id":"43d583213bfaa02f","type":"mqtt in","z":"c0e718067b85a8cc","g":"7b7be9e90c83eaff","name":"","topic":"Ambrogio/Robertino/pos","qos":"2","datatype":"auto-detect","broker":"6db118ed1b0c56de","nl":false,"rap":true,"rh":0,"inputs":0,"x":160,"y":830,"wires":[["2218191fe090798b","55b21eb55f283b61","36f55a2e222b2326"]]},{"id":"2218191fe090798b","type":"show-value","z":"c0e718067b85a8cc","g":"7b7be9e90c83eaff","name":"","path":"","x":160,"y":880,"wires":[[]]},{"id":"55b21eb55f283b61","type":"function","z":"c0e718067b85a8cc","g":"7b7be9e90c83eaff","name":"GPS Tracker position","func":"let xMax = 5.229248;\nlet xMin = 5.228394;\nlet yMax = 59.392793;\nlet yMin = 59.392350;\nconst numpos = 4000;\nlet arr = flow.get(\"Robertino\") || []\nlet arr2 = []\n//node.warn(msg.payload);\nlet data = msg.payload;\n//var temp=data.split(\",\");\nlet msg2 = {}\n//node.warn(data);\n\n// lon/lat har format ddmm.mmmmmm\nlet lat = Number(data.lat) / 100;\n//node.warn(lat)\nlet intpart = parseInt(lat);\nlet fractpart = lat - intpart;\nlat = intpart + (fractpart * 1.666667);\n//node.warn(lat)\n\nlet lon = Number(data.lon) / 100;\nintpart = parseInt(lon);\nfractpart = lon - intpart;\nlon = intpart + (fractpart * 1.666667);\n\nlet time = data.utc;\n//node.warn(lat);\n//node.warn(lon);\n\n// Ignorer verdier utenfor kartet\nif (lat < yMax && lat > yMin && lon < xMax && lon > xMin) arr.push({\"lat\": lat, \"lon\": lon, \"UTC\": time});\n\n// Remove old positions\nlet n = arr.length;\n//node.warn(n);\narr2 = arr.slice(n - numpos);\n\n//node.warn(arr);\n//node.warn(arr2);\nflow.set(\"Robertino\", arr2);\nmsg.payload = \" \";\nmsg2.topic = \"Robertino\";\nmsg2.payload = time;\nreturn [msg, msg2];","outputs":2,"timeout":0,"noerr":0,"initialize":"","finalize":"","libs":[],"x":410,"y":830,"wires":[["c466d510d0476f0e"],["8c73af56d7fdc323"]]},{"id":"8c73af56d7fdc323","type":"ui_text","z":"c0e718067b85a8cc","g":"7b7be9e90c83eaff","group":"a7f439d7b1e0bbd4","order":9,"width":"8","height":"1","name":"Robertino last track","label":"{{msg.topic}} last trck","format":"<font size=6>{{msg.payload}}<font size=3> UTC","layout":"row-spread","className":"","style":false,"font":"","fontSize":16,"color":"#000000","x":740,"y":870,"wires":[]},{"id":"36f55a2e222b2326","type":"debug","z":"c0e718067b85a8cc","g":"7b7be9e90c83eaff","name":"debug 180","active":true,"tosidebar":false,"console":false,"tostatus":true,"complete":"payload","targetType":"msg","statusVal":"","statusType":"counter","x":160,"y":930,"wires":[]},{"id":"c466d510d0476f0e","type":"function","z":"c0e718067b85a8cc","g":"7b7be9e90c83eaff","name":"Reset delay","func":"//msg.delay = 600000;\nvar m1 = {reset:true};\nreturn [[m1,msg]];","outputs":1,"noerr":0,"initialize":"","finalize":"","libs":[],"x":610,"y":820,"wires":[["93ac548a0bcda12d"]]},{"id":"93ac548a0bcda12d","type":"delay","z":"c0e718067b85a8cc","g":"7b7be9e90c83eaff","name":"","pauseType":"delay","timeout":"2","timeoutUnits":"seconds","rate":"1","nbRateUnits":"1","rateUnits":"second","randomFirst":"1","randomLast":"5","randomUnits":"seconds","drop":false,"allowrate":false,"outputs":1,"x":770,"y":820,"wires":[["3f628673e320b6ef"]]},{"id":"6db118ed1b0c56de","type":"mqtt-broker","name":"DaleMQTT","broker":"172.16.0.94","port":"1883","clientid":"34567890","autoConnect":true,"usetls":false,"protocolVersion":"4","keepalive":"60","cleansession":false,"autoUnsubscribe":true,"birthTopic":"","birthQos":"0","birthPayload":"","birthMsg":{},"closeTopic":"","closeQos":"0","closePayload":"","closeMsg":{},"willTopic":"","willQos":"0","willPayload":"","willMsg":{},"userProps":"","sessionExpiry":""},{"id":"a7f439d7b1e0bbd4","type":"ui_group","name":"Status","tab":"55b4bf40cf1e815d","order":2,"disp":true,"width":"8","collapse":false,"className":""},{"id":"55b4bf40cf1e815d","type":"ui_tab","name":"Ambrogios","icon":"android","order":23,"disabled":false,"hidden":false},{"id":"af4caa710bbb0c2a","type":"group","z":"c0e718067b85a8cc","name":"Presenter GPS data i chart","style":{"label":true},"nodes":["8e291ee416e96572","597d06bd03195022","6969dbab26c1048b","51404847151e1696","3f628673e320b6ef"],"x":24,"y":999,"w":742,"h":112},{"id":"8e291ee416e96572","type":"inject","z":"c0e718067b85a8cc","g":"af4caa710bbb0c2a","name":"","props":[{"p":"payload"},{"p":"topic","vt":"str"}],"repeat":"","crontab":"","once":true,"onceDelay":"10","topic":"","payload":"","payloadType":"date","x":140,"y":1050,"wires":[["3f628673e320b6ef"]]},{"id":"597d06bd03195022","type":"function","z":"c0e718067b85a8cc","g":"af4caa710bbb0c2a","name":"Read temp positions & parse","func":"//let poss = flow.get(\"temp\");\nlet poss = flow.get(\"Robertino\") || []\nvar data = []\nvar data0 = []\nvar data1 = []\nvar data2 = []\nvar data3 = []\nlet j = 0;\n//for (let i in poss.data[0].attributes.positions){\n// data0.push({x: poss.data[0].attributes.positions[i].longitude,y:poss.data[0].attributes.positions[i].latitude});\nfor (let i in poss){\n if (poss[i].lat > 59 && poss[i].lon > 5) data0.push({x: poss[i].lon,y:poss[i].lat});\n// node.warn(poss.data[0].attributes.positions[i].longitude);\nj = i;\n}\n// Ta vare på siste pos for oransje markering plott (nåværende pos)\n//data2.push({x: poss.data[0].attributes.positions[0].longitude,y:poss.data[0].attributes.positions[0].latitude});\n//data2.push({x: poss.data[0].attributes.positions[1].longitude,y:poss.data[0].attributes.positions[1].latitude});\ndata2.push({x: poss[j-3].lon,y:poss[j-3].lat});\ndata2.push({x: poss[j-2].lon,y:poss[j-2].lat});\ndata2.push({x: poss[j-1].lon,y:poss[j-1].lat});\ndata2.push({x: poss[j].lon,y:poss[j].lat});\n\n// Omriss av plenen\ndata1 = flow.get(\"plen\");\n// Omriss av elendom\ndata3 = flow.get(\"eiendom\");\n\n\ndata[1] = data1; // Lawnlimits\ndata[3] = data3; // Lawnlimits\ndata[2] = data0; // Track\ndata[0] = data2; // Last track\n//node.warn(data[2]);\n\nmsg.payload = [{\n \"series\": [\"A\",\"B\",\"C\",\"D\"],\n \"xAxisID\": 'custom-x-axis',\n \"yAxisID\": 'custom-y-axis',\n \"data\": data,\n \"labels\": [\"\"]\n}]\n\nreturn msg;\n","outputs":1,"timeout":0,"noerr":0,"initialize":"","finalize":"","libs":[],"x":450,"y":1040,"wires":[["51404847151e1696"]]},{"id":"6969dbab26c1048b","type":"function","z":"c0e718067b85a8cc","g":"af4caa710bbb0c2a","name":"modify chart options","func":"/*let centX = 5.1372;//2285;\nlet xW = 0.0005;\nlet xH = 0.0005;\nlet centY = 59.2355;//3924;\nlet xMax = centX + xW;\nlet xMin = centX - xW;\nlet yMax = centY + xH;\nlet yMin = centY - xH;\n*/\nlet xMax = 5.229248;\nlet xMin = 5.228394;\nlet yMax = 59.392793;\nlet yMin = 59.392350;\n\n\nlet gridcolors = 'rgba(255, 160, 0, 0.3)'\nmsg.ui_control = {\n options: {\n legend: {\n display: false\n },\n tooltips: {\n enabled: false\n },\n scales: {\n xAxes: [{\n type: 'linear',\n id: 'custom-x-axis',\n position:'bottom',\n padding:100,\n gridLines:{\n color:gridcolors,\n zeroLineColor:'rgba(123, 113, 113, 0.75)',\n tickMarkLength:7,\n drawTicks:false\n },\n ticks: {\n fontColor:\"#ccc\",\n max: xMax,\n min: xMin,\n stepSize: 0.0001\n \n }\n }],\n yAxes: [{\n id: 'custom-y-axis',\n \n gridLines:{\n color:gridcolors,\n zeroLineColor:'red',\n tickMarkLength:5,\n drawTicks:false\n },\n ticks: {\n fontColor:\"#ccc\",\n max: yMax,\n min: yMin,\n stepSize: 0.0001\n }\n }]\n }\n }\n}\ndelete msg.payload\nreturn msg;","outputs":1,"timeout":"","noerr":0,"initialize":"","finalize":"","libs":[],"x":430,"y":1070,"wires":[["51404847151e1696"]]},{"id":"51404847151e1696","type":"ui_chart","z":"c0e718067b85a8cc","g":"af4caa710bbb0c2a","name":"Plen","group":"7ab0161dbb92ab3f","order":3,"width":"30","height":"20","label":"","chartType":"line","legend":"false","xformat":"auto","interpolate":"linear","nodata":"","dot":false,"ymin":"","ymax":"","removeOlder":"1","removeOlderPoints":"","removeOlderUnit":"3600","cutout":0,"useOneColor":false,"useUTC":false,"colors":["#ff7f00","#00ff00","#8585fa","#2ca02c","#98df8a","#d62728","#ff9896","#9467bd","#c5b0d5"],"outputs":1,"useDifferentColor":false,"className":"","x":690,"y":1040,"wires":[[]]},{"id":"3f628673e320b6ef","type":"junction","z":"c0e718067b85a8cc","g":"af4caa710bbb0c2a","x":280,"y":1050,"wires":[["597d06bd03195022","6969dbab26c1048b"]]},{"id":"7ab0161dbb92ab3f","type":"ui_group","name":"Map","tab":"55b4bf40cf1e815d","order":3,"disp":true,"width":"32","collapse":false,"className":""},{"id":"f51cf63d67a5a2e5","type":"group","z":"c0e718067b85a8cc","name":"Les geojson fil og lagre koordinater i flow.plen","style":{"label":true},"nodes":["7a8aba7d1b2233c3","c3ac6155687bd8b2","7a3a31186dd532c5","65861ab97117862e"],"x":24,"y":1139,"w":1012,"h":82},{"id":"7a8aba7d1b2233c3","type":"json","z":"c0e718067b85a8cc","g":"f51cf63d67a5a2e5","name":"","property":"payload","action":"","pretty":false,"x":670,"y":1180,"wires":[["c3ac6155687bd8b2"]]},{"id":"c3ac6155687bd8b2","type":"function","z":"c0e718067b85a8cc","g":"f51cf63d67a5a2e5","name":"Plukk ut pos fra fil og lagre i flow","func":"var plen = []\nvar eiendom = []\n\nfor (var i in msg.payload.features[0].geometry.coordinates){\n let tempobj = {x: msg.payload.features[0].geometry.coordinates[i][0] ,y: msg.payload.features[0].geometry.coordinates[i][1]};\n plen.push(tempobj);\n}\n//payload.features[1].geometry.coordinates[0][0]\n//payload.features[1].geometry.coordinates[0][9]\n//payload.features[1].geometry.coordinates[0][9][0]\n//payload.features[1].geometry.coordinates[0][9][1]\nfor (var i in msg.payload.features[1].geometry.coordinates[0]) {\n let tempobj = { x: msg.payload.features[1].geometry.coordinates[0][i][0], y: msg.payload.features[1].geometry.coordinates[0][i][1] };\n eiendom.push(tempobj);\n}\n//node.warn(plen);\nflow.set(\"plen\", plen);\nflow.set(\"eiendom\", eiendom);\nreturn msg;","outputs":1,"timeout":0,"noerr":0,"initialize":"","finalize":"","libs":[],"x":880,"y":1180,"wires":[["3f628673e320b6ef"]]},{"id":"7a3a31186dd532c5","type":"file in","z":"c0e718067b85a8cc","g":"f51cf63d67a5a2e5","name":"","filename":"/home/nodered/PlenV4Eiendom.geojson","filenameType":"str","format":"utf8","chunk":false,"sendError":false,"encoding":"none","allProps":false,"x":390,"y":1180,"wires":[["7a8aba7d1b2233c3"]]},{"id":"65861ab97117862e","type":"inject","z":"c0e718067b85a8cc","g":"f51cf63d67a5a2e5","name":"","props":[{"p":"payload"},{"p":"topic","vt":"str"}],"repeat":"","crontab":"","once":false,"onceDelay":0.1,"topic":"","payload":"","payloadType":"date","x":130,"y":1180,"wires":[["7a3a31186dd532c5"]]}] uPython programmet: import ubinascii import machine from umqtt.simple import MQTTClient from machine import UART import network, utime, machine import os import uasyncio import json uart = UART(2, 9600) SSID = "WLAN-SSID" SSID_PASSWORD = "WLAN-PWD" filenum = 0 SERVER = "172.16.0.94" CLIENT_ID = ubinascii.hexlify(machine.unique_id()) TOPIC = "Ambrogio/Robertino/pos" gps_target = "$GPGGA" gps_target = gps_target.encode('ASCII') async def readAndSend(): global filenum while True: await uasyncio.sleep_ms(1000 * 30) #Run every 30 sec sta_if = network.WLAN(network.STA_IF) if not sta_if.isconnected(): print('connecting to network...') sta_if.active(True) sta_if.connect(SSID, SSID_PASSWORD) while not sta_if.isconnected(): print("Attempting to connect....") await uasyncio.sleep_ms(1000) print('Connected! Network config:', sta_if.ifconfig()) mqttClient = MQTTClient(CLIENT_ID, SERVER, keepalive=60) mqttClient.connect() print(f"Connected to MQTT Broker: {SERVER}") locfilenum = filenum # Store current filenumber filenum = filenum + 1 # New filenumber for logging filename = 'data'+str(locfilenum)+'.txt' if filename in os.listdir(): print('Found '+ filename) f = open(filename, 'r') line = f.readline() while line != "": print(filename+", MQTT:"+line.strip()) mqttClient.publish(TOPIC, str(line)) await uasyncio.sleep_ms(500) line = f.readline() print("file finished") print("Removing ", filename) os.remove(filename) mqttClient.disconnect() print("Disconnect from wlan") sta_if.disconnect() def left(s, amount): return s[:amount] async def main(): global filenum print("Cleanup old files...") # Delete all existing data files on startup for f in os.listdir(): if left(f, 4) == "data": print("Deleting: ", f) os.remove(f) print("All data files deleted.") while True: if uart.any() > 0: await uasyncio.sleep_ms(100) gps_msg = uart.readline() if len(gps_msg) >= 60: gps_type = left(gps_msg, 6) if gps_type == gps_target: utc = gps_msg[7:13] lat = gps_msg[17:27] lon = gps_msg[30:41] newmsg = {} newmsg['utc'] = utc newmsg['lat'] = lat newmsg['lon'] = lon jmsg = json.dumps(newmsg) # Create JSON object for datafile #if lat >= '5923' and lat <= '5924' and lon >= '513' and lon <= '514': filename = 'data'+str(filenum)+'.txt' print(filename+", GPS:"+jmsg) f = open(filename, 'a') f.write(jmsg) f.write("\r\n") f.close() await uasyncio.sleep_ms(200) # Start hele sulamitten event_loop = uasyncio.get_event_loop() event_loop.create_task(main()) event_loop.create_task(readAndSend()) event_loop.run_forever() ESP32 forsynes pt via USB fra en batteribank og GPS forsynes fra Vin og GND pinnene på ESP32. GPS-TX går til ESP-RX2, GPS-RX til ESP-TX2. Har sett i minst ett eksempel på nett at NEO6 GPSen forsynes fra 3.3V men spec sier at NEO6 skal ha 3.7-5V. Prøvde først med 3.3V og det virket men fikk en del mangelfulle setninger fra GPS så med 5V gikk det myyyyye bedre... Så langt ha jeg brukt en 10Ah batteribank og den holder liv i greiene i mange døgn men skal hacke meg inn på strømforsyningen i klipperen etterhvert...

Har i noen måneder brukt uike verktøy til å enkelt generere yaml kode kun ved å beskrive behovene mine. Blir bare mer og mer imponert over utviklingen og hva det er mulig å få til - dersom spørsmålene stilles riktig - med høyst begrensede programmeringskunnskaper. Et eksempel er denne som fargekoder timene basert på rangering av strømprisen: type: custom:apexcharts-card series: - entity: sensor.nordpool_kwh_oslo_nok_3_10_0 name: Pris type: column float_precision: 4 data_generator: > const today = entity.attributes.raw_today; const tomorrow = entity.attributes.raw_tomorrow; const allData = today.concat(tomorrow); const categorizeHours = (data) => { const sorted = [...data].sort((a, b) => a.value - b.value); const categories = { lowestFour: sorted.slice(0, 4).map(p => p.start), nextFour: sorted.slice(4, 8).map(p => p.start), middle: sorted.slice(8, 16).map(p => p.start), nextHighestFour: sorted.slice(16, 20).map(p => p.start), highestFour: sorted.slice(20).map(p => p.start) }; return categories; }; const todayCategories = categorizeHours(today); const tomorrowCategories = categorizeHours(tomorrow); const getColor = (start) => { for (const day of [todayCategories, tomorrowCategories]) { if (day.lowestFour.includes(start)) return '#006400'; // Dark Green if (day.nextFour.includes(start)) return '#90EE90'; // Light Green if (day.middle.includes(start)) return '#1E90FF'; // Blue if (day.nextHighestFour.includes(start)) return '#FFA500'; // Orange if (day.highestFour.includes(start)) return '#FF0000'; // Red } return '#808080'; // Default Gray (shouldn't occur) }; return allData.map(p => ({ x: new Date(p.start).getTime(), y: p.value, fillColor: getColor(p.start) })); now: show: true label: Nå graph_span: 2d span: start: day locale: nb apex_config: chart: height: 300px zoom: enabled: true toolbar: show: true tools: zoom: true zoomin: false zoomout: false pan: false reset: true legend: show: false dataLabels: enabled: false xaxis: type: datetime labels: datetimeFormatter: month: ddd day: ddd yaxis: forceNiceScale: false min: 0 plotOptions: bar: columnWidth: '95%' På den måten kan jeg se når f.eks. varmekabler på bad og VVB er aktvert (som begge er aktivert på døgnets 8 billigste timer, altså mørkegrønne og lysegrønne farger). Etter nærmere ettertanke kan VVB og varmekabler uansett være aktivert om prisen er under 0.2 kroner. Gir derfor AI-assistenten følgende beskjed: can you modify the code and make all bars dark green if the value is bellow 0.2? og jeg får dette resultatet: Answer Certainly! I'll modify the code to make all bars dark green if their value is below 0.2, while keeping the color categorization for the rest. Here's the updated version: type: custom:apexcharts-card series: - entity: sensor.nordpool_kwh_oslo_nok_3_10_0 name: Pris type: column float_precision: 4 data_generator: > const today = entity.attributes.raw_today; const tomorrow = entity.attributes.raw_tomorrow; const allData = today.concat(tomorrow); const categorizeHours = (data) => { const sorted = [...data].sort((a, b) => b.value - a.value); // Sort descending const categories = { highestFour: sorted.slice(0, 4).map(p => p.start), nextHighestFour: sorted.slice(4, 8).map(p => p.start), middle: sorted.slice(8, 16).map(p => p.start), nextLowestFour: sorted.slice(16, 20).map(p => p.start), lowestFour: sorted.slice(20).map(p => p.start) }; return categories; }; const todayCategories = categorizeHours(today); const tomorrowCategories = categorizeHours(tomorrow); const getColor = (start, value) => { if (value < 0.2) return '#006400'; // Dark Green for values below 0.2 for (const day of [todayCategories, tomorrowCategories]) { if (day.highestFour.includes(start)) return '#FF0000'; // Red if (day.nextHighestFour.includes(start)) return '#FFA500'; // Orange if (day.middle.includes(start)) return '#1E90FF'; // Blue if (day.nextLowestFour.includes(start)) return '#90EE90'; // Light Green if (day.lowestFour.includes(start)) return '#00FF00'; // Green } return '#808080'; // Default Gray (shouldn't occur) }; return allData.map(p => ({ x: new Date(p.start).getTime(), y: p.value, fillColor: getColor(p.start, p.value) })); now: show: true label: Nå graph_span: 2d span: start: day locale: nb apex_config: chart: height: 300px zoom: enabled: true toolbar: show: true tools: zoom: true zoomin: false zoomout: false pan: false reset: true legend: show: false dataLabels: enabled: false xaxis: type: datetime labels: datetimeFormatter: month: ddd day: ddd yaxis: forceNiceScale: false min: 0 plotOptions: bar: columnWidth: '95%' Som gir meg denne: ...så kan jeg lage automasjonene tilsvarende. Andre som har fått tilsvarende hjelp?

Ganske utruleg at eg ikkje kopla, men eg har fått samme problemet sjøl. Merka det nok ikkje sjøl då vi dreiv med oppussing av stove og kjøkken der stemmestyring via Google Home blir brukt. Trur det må ha oppstått når det vart innført betaling for å ha Google Home integrert. Har funne ei løysing som fungerer greit - opprett ein virtuell device per event som skal køyrast, aktiver voice command på denne og trigg aktuelle events når virtuell device blir aktivert. Det burde være unødvendig, men det fungerer...