Therese With Berge

Research Scientist

(+47) 922 93 927
therese.berge@nibio.no

Place
Ås H7

Visiting address
Høgskoleveien 7, 1433 Ås

Biography

My research topics:
- Weed management with reduced environmental impact
- Precision farming and robotics
- Integrated weed management in field vegetables
- Weed seed predation (model weed species: Echinochloa crus-galli, Chenopodium album)
- Alien invasive plants (Impatiens balsamifera)

Read more
To document

Abstract

Vegetables and other row-crops represent a large share of the agricultural production. There is a large variation in crop species, and a limited availability in specialized herbicides. The robot presented here utilizes systematic growing techniques to navigate and operate in the field. By the use of machine vision it separates seeded vegetable crops from weed. Each weed within the row is treated with individual herbicide droplets, without affecting the crop. This results in a significant reduction in herbicide use, and allows for the use of herbicides that would otherwise harm the crop. The robot is tailored to this purpose with cost, maintainability, efficient operation and robustness in mind. The three-wheeled design is unconventional, and the design maintains maneuverability and stability with the benefit of reduced weight, complexity and cost. Indoor pot trials with four weed species demonstrated that the Drop-on-Demand system (DoD) could control the weeds with as little as 7.6 μg glyphosate or 0.15 μg iodosulfuron per plant. The results also highlight the importance of liquid characteristics for droplet stability and leaf retention properties. The common herbicide glyphosate had no effect unless mixed with suitable additives. A field trial with the robot was performed in a carrot field, and all the weeds were effectively controlled with the DoD system applying 5.3 μg of glyphosate per droplet. The robot and DoD system represent a paradigm shift to the environmental impact and health risks of weed control, while providing a valuable tool to the producers.

Abstract

Creeping perennial weeds are of major concern in organically grown cereals. In the present study, the effects of different timing of mouldboard ploughing with or without a preceding stubble cultivation period, on weeds and spring cereals were studied. The experiments were conducted at two sites in Norway during a two and three-year period, respectively, with the treatments repeated on the same plots. The soil cultivation treatments were a stubble disc-harrowing cultivation period followed by mouldboard ploughing and only mouldboard ploughing. The timing of the treatments were autumn or spring. The density and biomass of the aboveground shoots of Cirsium arvense (L.) Scop., Elymus repens (L.) Gould, Sonchus arvensis L. and Stachys palustris L. as well as the total aboveground biomass of the spring cereal crop (oats) were assessed. The control efficiency of C. arvense and S. arvensis was closely related to timing of the cultivation treatments. Cultivation in spring decreased the population of C. arvense and S. arvensis compared to autumn cultivation. For E. repens, timing of the treatments had no significant effect: the important factor was whether stubble cultivation was carried out (best control) or not. The overall best strategy for controlling the present perennial weed population was stubble cultivation followed by ploughing in spring. However, the associated relative late sowing of the spring cereal crop and lowered crop biomass, were important drawbacks.

To document

Abstract

The success of precision agriculture relies largely on our ability to identify how the plants’ growth limiting factors vary in time and space. In the field, several stress factors may occur simultaneously, and it is thus crucial to be able to identify the key limitation, in order to decide upon the correct contra-action, e.g., herbicide application. We performed a pot experiment, in which spring wheat was exposed to water shortage, nitrogen deficiency, weed competition (Sinapis alba L.) and fungal infection (Blumeria graminis f. sp. tritici) in a complete, factorial design. A range of sensor measurements were taken every third day from the two-leaf stage until booting of the wheat (BBCH 12 to 40). Already during the first 10 days after stress induction (DAS), both fluorescence measurements and spectral vegetation indices were able to differentiate between non-stressed and stressed wheat plants exposed to water shortage, weed competition or fungal infection. This meant that water shortage and fungal infection could be detected prior to visible symptoms. Nitrogen shortage was detected on the 11–20 DAS. Differentiation of more than one stress factors with the same index was difficult.

To document

Abstract

Arable weeds are generally distributed in patches, while herbicides are applied uniformly. Herbicides can be saved if only the patches are sprayed, i.e. patch spraying (PS). Bottlenecks for cost-effective PS are weed monitoring technology and valid technology-based decision rules for PS (thresholds). The novel machine vision algorithm Weedcer has been developed as an efficient weed monitoring tool for PS. Weedcer estimates the proportions of young weed leaves and cereal leaves in high resolution red–green–blue images. We conducted field trials to test relative weed cover (RWC) and relative mayweed cover (RMC) estimated by Weedcer as decision rules for PS. RWC is the total weed cover divided by the total plant cover and RMC is the mayweed cover divided by the total plant cover. The main criterion for evaluation and basis of these thresholds was the measured grain yield. Images (about 0.06-m2) were acquired with a GPS guided autonomous field robot in spring, the normal time for spraying seed-propagated broadleaf weeds in both winter – and spring cereals in Norway. Three map-based trials (weed monitoring and spraying in two separate operations) showed that mean RWC per management unit (12.0 × 12.5-m) was generally adequate. In winter wheat heavily infested with scentless mayweed (Tripleurospermum inodorum (L.) Sch.Bip.) and/or scented mayweed (Matricaria recutita L.), the mean RMC per management unit was more adequate. Progress during the project allowed three additional trials conducted in real-time (weed monitoring and spraying in the same operation). These were conducted with the robot in spring cereals, and showed that a weighted moving average of RWC per image was adequate. The sprayed and unsprayed management units in these trials were minimum 3.0 × 3.0-m and 0.5 × 3.0-m, respectively. Results indicated that the Weedcer-based thresholds should be lower in wheat (Triticum aestivum) than in barley (Hordeum vulgare).

To document

Abstract

A possible cost-effective real-time patch spraying implementation against seed-propagated broad-leaved weeds in cereals is a camera mounted in front of the tractor taking images at feasible distances in the direction of travel, on-board image analysis software and entire boom switched on and off. To assess this implementation, manual weed counts (0.25 m(2) quadrats) in a 1.5 m x 2 m grid, were used to simulate camera outputs. Each quadrat was classified into 'spray' and 'not spray' decisions based on a threshold model, and the resulting map defined the 'ground truth'. Subsequently, 'on/off' spraying at larger control areas where sizes were given by the boom width and image distance, and spraying decision controlled by weed status at the single quadrat simulating the camera's view, were simulated. These coarser maps were compared with 'ground truth', to estimate mapping error (area above threshold not sprayed), spraying error (area below threshold sprayed), total error (sum of mapping and spraying error) and the herbicide reduction. Three levels of the threshold model were tested. Results were used to fit models that predict errors from boom width and image distance. Size of control area did not on average affect the magnitude of the simulated herbicide reductions, but the bigger the control area the higher the risk that the simulated herbicide reduction deviate from the reduction in 'ground truth'. Mean simulated herbicide reductions were 42-59%, depending on threshold level. Only minor differences due to threshold level were seen for mean mapping and spraying errors at given spraying resolutions. Using original threshold level and image distance 2 m, predicted total errors for boom widths 2 m, 6 m, 20 m and 40 m would be 6%, 10%, 16% and 17%, respectively. Results indicate that control area should not exceed about 10 m 2 if acceptable total error is maximum 10%.

To document

Abstract

Docks (Rumex spp.) are a considerable problem in grassland production worldwide. We investigated how different cultural management techniques affected dock populations during grassland renewal: (I) renewal time, (II) companion crop, (III) false seedbed, (IV) taproot cutting (V), plough skimmer and (VI) ploughing depth. Three factorial split-split plot experiments were carried out in Norway in 2007–2008 (three locations), 2008–2009 (one location) and 2009 (one location). After grassland renewal, more dock plants emerged from seeds than from roots. Summer renewal resulted in more dock seed and root plants than spring renewal. Adding a spring barley companion crop to the grassland crop often reduced dock density and biomass. A false seedbed resulted in 71% fewer dock seed plants following summer renewal, but tended to increase the number of dock plants after spring renewal. In some instances, taproot cutting resulted in less dock biomass, but the effect was weak and inconsistent, and if ploughing was shallow (16 cm) or omitted, it instead increased dock root plant emergence. Fewer root plants emerged after deep ploughing (24 cm) compared to shallow ploughing, and a plough skimmer tended to reduce the number further. We conclude that a competitive companion crop can assist in controlling both dock seed and root plants, but it is more important that the renewal time is favourable to the main crop. Taproot cutting in conjunction with ploughing is not an effective way to reduce dock root plants, but ploughing is more effective if it is deep and a skimmer is used.

To document

Abstract

Vegetables and other row-crops represent a large share of the agricultural production. There is a large variation in crop species, and a limited availability in specialized herbicides. The robot presented here utilizes systematic growing techniques to navigate and operate in the field. By the use of machine vision it separates seeded vegetable crops from weed. Each weed within the row is treated with individual herbicide droplets, without affecting the crop. This results in a significant reduction in herbicide use, and allows for the use of herbicides that would otherwise harm the crop. The robot is tailored to this purpose with cost, maintainability, efficient operation and robustness in mind. The three-wheeled design is unconventional, and the design maintains maneuverability and stability with the benefit of reduced weight, complexity and cost. Indoor pot trials with four weed species demonstrated that the Drop-on-Demand system (DoD) could control the weeds with as little as 7.6 μg glyphosate or 0.15 μg iodosulfuron per plant. The results also highlight the importance of liquid characteristics for droplet stability and leaf retention properties. The common herbicide glyphosate had no effect unless mixed with suitable additives. A field trial with the robot was performed in a carrot field, and all the weeds were effectively controlled with the DoD system applying 5.3 μg of glyphosate per droplet. The robot and DoD system represent a paradigm shift to the environmental impact and health risks of weed control, while providing a valuable tool to the producers.

Abstract

Creeping perennial weeds are of major concern in organically grown cereals. In the present study, the effects of different timing of mouldboard ploughing with or without a preceding stubble cultivation period, on weeds and spring cereals were studied. The experiments were conducted at two sites in Norway during a two and three-year period, respectively, with the treatments repeated on the same plots. The soil cultivation treatments were a stubble disc-harrowing cultivation period followed by mouldboard ploughing and only mouldboard ploughing. The timing of the treatments were autumn or spring. The density and biomass of the aboveground shoots of Cirsium arvense (L.) Scop., Elymus repens (L.) Gould, Sonchus arvensis L. and Stachys palustris L. as well as the total aboveground biomass of the spring cereal crop (oats) were assessed. The control efficiency of C. arvense and S. arvensis was closely related to timing of the cultivation treatments. Cultivation in spring decreased the population of C. arvense and S. arvensis compared to autumn cultivation. For E. repens, timing of the treatments had no significant effect: the important factor was whether stubble cultivation was carried out (best control) or not. The overall best strategy for controlling the present perennial weed population was stubble cultivation followed by ploughing in spring. However, the associated relative late sowing of the spring cereal crop and lowered crop biomass, were important drawbacks.

Abstract

With the Directive 2009/128/EC on sustainable use of pesticides, reductions in herbicide use is a European target. The aim of this study was to compare the fi eld-specifi c herbicide use resulting from simulated integrated weed management (IWM) with farmer’s actual use. Two IWM tools applicable for cereals were explored: VIPS – a web-based decision support system, and DAT sensor – a precision farming technology for patch spraying. VIPS (adaptation of Danish “Crop Protection Online”) optimizes herbicide – and dose to weed species densityand growth stage (including ALS-herbicide resistant populations), temperature, expected yield, cereal species- and growth stage. Weeds were surveyed (0.25 m2, n=23-31) prior to post-emergence spraying in spring 2013 (six fi elds) and 2014 (eight fi elds). DAT sensor enables automatic patch spraying of annual weeds within cereals. It consists of an RGB camera and custom-made image analysis. DAT sensor acquired more than 900 images (0.06 m2) per fi eld. Threshold for simulated patch spraying was relative weed cover (weed cover/ total vegetation cover) = 0.042. Treatment frequency index (TFI, actual dose/maximum approved dose summed for all herbicides) was calculated. Without resistance strategy, average TFI for VIPS was higher for winter wheat (0.96) than for spring cereals (0.38). Spring cereal fi elds with resistance strategies gave an average TFI of 1.45. Corresponding TFI for farmer’s applications were 1.40, 0.90 and 1.26, respectively. For one fi eld wherein both tools were explored in 2013 and 2014, TFI values for VIPS were 1.86 and 1.50 due to resistant Stellaria media, while TFI for farmer’s sprayings were around 1.00. DAT sensor simulated herbicide savings of 69% and 99%, corresponding to TFI values of 0.58 and 0.01, respectively. As measured by TFI, DAT sensor showed a higher potential in herbicide savings than VIPS. VIPS is available without costs to end-users today, while DAT sensor represents a future tool.

To document

Abstract

The success of precision agriculture relies largely on our ability to identify how the plants’ growth limiting factors vary in time and space. In the field, several stress factors may occur simultaneously, and it is thus crucial to be able to identify the key limitation, in order to decide upon the correct contra-action, e.g., herbicide application. We performed a pot experiment, in which spring wheat was exposed to water shortage, nitrogen deficiency, weed competition (Sinapis alba L.) and fungal infection (Blumeria graminis f. sp. tritici) in a complete, factorial design. A range of sensor measurements were taken every third day from the two-leaf stage until booting of the wheat (BBCH 12 to 40). Already during the first 10 days after stress induction (DAS), both fluorescence measurements and spectral vegetation indices were able to differentiate between non-stressed and stressed wheat plants exposed to water shortage, weed competition or fungal infection. This meant that water shortage and fungal infection could be detected prior to visible symptoms. Nitrogen shortage was detected on the 11–20 DAS. Differentiation of more than one stress factors with the same index was difficult.

Abstract

In Europe there is an on-going process on implementing regulations aimed at reducing pollution from agricultural production systems, i.e. the Water Framework Directive and the Framework Directive for Sustainable Use of Pesticides. At the same time, there is an increasing focus on food security possibly leading to continued intensification of agricultural production with increased use of external inputs, such as pesticides and fertilizers. Application of sustainable production systems can only be achieved if they balance conflicting environmental and economic effects. In Norway, cereal production is of large importance for food security and reduction of soil and phosphorus losses, as well as pesticide use and leaching/runoff in the cereal production are of special concern. Therefore, we need to determine the most sustainable and effective strategies to reduce loss of top soil, phosphorus and pesticides while maintaining cereal yields. A three-year research project, STRAPP, is addressing these concerns. A catchment area dominated by cereal production is our common research arena within STRAPP. Since 1992 a database (JOVA) with data for soil erosion, nutrient and pesticide leaching/runoff (i.e. concentrations in stream water), yield, and agricultural management practices (fertilization, use of pesticides, soil tillage and rotations) has been established for this catchment allowing us to compare a unique diversity in cropping strategies in a defined location. An important part of STRAPP focuses on developing ‘best plant protection strategies’ for cereal fields in the study area, based on field inventories (manual and sensor based) of weeds and common diseases, available forecast systems, and pesticide leaching risk maps. The results of field studies during the growing seasons of 2013 and 2014 will be presented, with a focus on possible integrated pest management (IPM) strategies for weeds and fungal diseases in cereal production. We will also present the project concept and methods for coupling optimized plant protection strategies to (i) modelling of phosphorus and pesticide leaching/runoff, as well as soil loss, and (ii) farm-economic impacts and adaptations. Further, methods for balancing the conflicting environmental and economic effects of the above practices, and the evaluation of instruments for increased adoption of desirable management practices will be outlined.

To document

Abstract

Vehicles which operate in agricultural row crops, need to strictly follow the established wheel tracks. Errors in navigation where the robot sways of its path with one or more wheels may damage the crop plants. The specific focus of this paper is on an agricultural robot operation in row cultures. The robot performs machine vision detecting weeds within the crop rows and treats the weeds by high precision drop-on-demand application of herbicide. The navigation controller of the robot needs to follow the established wheel tracks and minimize the camera system offset from the seed row. The problem has been formulated as a Nonlinear Model Predictive Control (NMPC) problem with the objective of keeping the vision modules centered over the seed rows, and constraining the wheel motion to the defined Wheel tracks. The system and optimization problem has been implemented in Python using the Casadi framework. The implementation has been evaluated through simulations of the system, and compared with a PD controller. The NMPC approach display advantages and better performance when facing the path constraints of operating in row crops.

To document

Abstract

Vehicles which operate in agricultural row crops, need to strictly follow the established wheel tracks. Errors in navigation where the robot sways of its path with one or more wheels may damage the crop plants. The specific focus of this paper is on an agricultural robot operation in row cultures. The robot performs machine vision detecting weeds within the crop rows and treats the weeds by high precision drop-on-demand application of herbicide. The navigation controller of the robot needs to follow the established wheel tracks and minimize the camera system offset from the seed row. The problem has been formulated as a Nonlinear Model Predictive Control (NMPC) problem with the objective of keeping the vision modules centered over the seed rows, and constraining the wheel motion to the defined Wheel tracks. The system and optimization problem has been implemented in Python using the Casadi framework. The implementation has been evaluated through simulations of the system, and compared with a PD controller. The NMPC approach display advantages and better performance when facing the path constraints of operating in row crops.

To document

Abstract

The Asterix project develops an autonomous robot for automatic weed control in row-crops. The system will only apply a fraction of the herbicide used in conventional application. The Food and Agriculture Organization of the United Nations estimate that the worlds food production needs to increase by 60 % by to feed the growing world population. This cannot be ful lled by conventional agricultural methods on the worlds available farm lands. Precision Agriculture (PA) is the concept of measuring eld variability and introducing this information as a feed-back to the crop management. PA can increase yields and optimize the resource inputs, and reduce environmental damages by avoiding excess use of herbicides, pesticides and fertilizers. The Asterix project is an ultra-precise weed control approach, where individual weed leaves are controlled by herbicide droplets. The droplets are dispensed by an 18 cm wide array of drop-on-demand nozzles (DoD).

To document

Abstract

Arable weeds are generally distributed in patches, while herbicides are applied uniformly. Herbicides can be saved if only the patches are sprayed, i.e. patch spraying (PS). Bottlenecks for cost-effective PS are weed monitoring technology and valid technology-based decision rules for PS (thresholds). The novel machine vision algorithm Weedcer has been developed as an efficient weed monitoring tool for PS. Weedcer estimates the proportions of young weed leaves and cereal leaves in high resolution red–green–blue images. We conducted field trials to test relative weed cover (RWC) and relative mayweed cover (RMC) estimated by Weedcer as decision rules for PS. RWC is the total weed cover divided by the total plant cover and RMC is the mayweed cover divided by the total plant cover. The main criterion for evaluation and basis of these thresholds was the measured grain yield. Images (about 0.06-m2) were acquired with a GPS guided autonomous field robot in spring, the normal time for spraying seed-propagated broadleaf weeds in both winter – and spring cereals in Norway. Three map-based trials (weed monitoring and spraying in two separate operations) showed that mean RWC per management unit (12.0 × 12.5-m) was generally adequate. In winter wheat heavily infested with scentless mayweed (Tripleurospermum inodorum (L.) Sch.Bip.) and/or scented mayweed (Matricaria recutita L.), the mean RMC per management unit was more adequate. Progress during the project allowed three additional trials conducted in real-time (weed monitoring and spraying in the same operation). These were conducted with the robot in spring cereals, and showed that a weighted moving average of RWC per image was adequate. The sprayed and unsprayed management units in these trials were minimum 3.0 × 3.0-m and 0.5 × 3.0-m, respectively. Results indicated that the Weedcer-based thresholds should be lower in wheat (Triticum aestivum) than in barley (Hordeum vulgare).

To document

Abstract

Lack of automatic weed detection tools has hampered the adoption of site-specific weed control in cereals. An initial object-oriented algorithm for the automatic detection of broad-leaved weeds in cereals developed by SINTEF ICT (Oslo, Norway) was evaluated. The algorithm ("WeedFinder") estimates total density and cover of broad-leaved weed seedlings in cereal fields from near-ground red-green-blue images. The ability of "WeedFinder" to predict 'spray'/'no spray' decisions according to a previously suggested spray decision model for spring cereals was tested with images from two wheat fields sown with the normal row spacing of the region, 0.125 m. Applying the decision model as a simple look-up table, "WeedFinder" gave correct spray decisions in 65-85% of the test images. With discriminant analysis, corresponding mean rates were 84-90%. Future versions of "WeedFinder" must be more accurate and accommodate weed species recognition.

To document

Abstract

A possible cost-effective real-time patch spraying implementation against seed-propagated broad-leaved weeds in cereals is a camera mounted in front of the tractor taking images at feasible distances in the direction of travel, on-board image analysis software and entire boom switched on and off. To assess this implementation, manual weed counts (0.25 m(2) quadrats) in a 1.5 m x 2 m grid, were used to simulate camera outputs. Each quadrat was classified into 'spray' and 'not spray' decisions based on a threshold model, and the resulting map defined the 'ground truth'. Subsequently, 'on/off' spraying at larger control areas where sizes were given by the boom width and image distance, and spraying decision controlled by weed status at the single quadrat simulating the camera's view, were simulated. These coarser maps were compared with 'ground truth', to estimate mapping error (area above threshold not sprayed), spraying error (area below threshold sprayed), total error (sum of mapping and spraying error) and the herbicide reduction. Three levels of the threshold model were tested. Results were used to fit models that predict errors from boom width and image distance. Size of control area did not on average affect the magnitude of the simulated herbicide reductions, but the bigger the control area the higher the risk that the simulated herbicide reduction deviate from the reduction in 'ground truth'. Mean simulated herbicide reductions were 42-59%, depending on threshold level. Only minor differences due to threshold level were seen for mean mapping and spraying errors at given spraying resolutions. Using original threshold level and image distance 2 m, predicted total errors for boom widths 2 m, 6 m, 20 m and 40 m would be 6%, 10%, 16% and 17%, respectively. Results indicate that control area should not exceed about 10 m 2 if acceptable total error is maximum 10%.