Autonomous weeders for Christmas tree plantations - a feasibility study
Bilag C
General motivation of using autonomous vehicle systems
Spyros Fountas and Henrik Have
The Royal Veterinary and Agricultural University, Section for AgroTechnology
Agriculture, horticulture and forestry have in the past benefited from a succession of technological developments that have brought greater productivity and economic efficiency. Historically, the emphasis of these developments has been on the mechanisation of operations to increase work rates achievable by individual operators through the use of larger, more powerful, machines. Today, however, the newer information based technologies seems to enable reliable autonomous field operations to be viable (Earl, 2000). This may make small-scale machinery an interesting alternative.
There are many reasons to justify the use of small autonomous vehicles. The main one may be to replace the heavy big tractors and machinery, which may damages the soil, and moreover requires large energy inputs. According to Grath (1997) light machinery compared to the conventional have advantages in:
 | Lowered soil compaction |
| Lower energy consumption |
| Less influence from weather |
| Increased yield |
| Increased permeability for water |
| Less risk for prolonged periods with surface water |
| Increased quality of seed bed. |
1.1 Vehicle efficiency and costs
Autonomous machines differ from present machinery in many ways. They can work for longer hours than manned machinery, they are less susceptible to weather and soil conditions and may have better utilization, which all together have considerable effects on costs and environmental consequences.
1.1.1 Machinery Utilisation
Operation of common manned machinery is in most cases restricted by labour availability (often only 5-6 hours a day on average), unsuitable weather, soil and crop conditions (up to 60 % of the time) and machinery down time (about 3 %). This means that presently used machinery only can be operated about 10-20 % of the total time.
If the labour restriction to a large extend can be removed this means that machinery utilisation may be increased up to 3-4 times. Additionally, if an autonomous machine is less sensitive to adverse weather conditions we can further increase the machinery utilisation further.
1.1.1.1 Machinery costs
A survey by Poulsen et al. (1997) of 500 full time farms in Denmark showed that on average the largest part of the total labour and machinery costs was the labour costs (31%), followed by machinery depreciation (21%), interest (18%), and maintenance (16%) and energy costs (5%). A shift to autonomous machines would give a major reduction in the labour costs and also a consider reduction because of better machine utilisation. However, high initial investment costs of autonomous systems may work in the other direction.
1.1.1.2 Operator limitations
Often the rate and quality of fieldwork are limited by the skill of the operator, as he has to make steering adjustments continuously while maintaining the attached implement at some level of acceptable performance. Although advances have been made in tractor and cab design to improve operator convenience and comfort, increased speeds, power and machine widths require the operator to be more attentive to the driving function, often at the expense of reduced equipment performance. The need to relieve him is perhaps the most frequently cited reason for the need of a vehicle guidance system (Wilson, 2000).
Kondo and Ting, (1998), states that even though many agricultural operations have been mechanized, there are still many laborious and monotonous tasks that are not suited for human beings, but require some intelligence to perform. Additionally, the availability of the farming workforce is decreasing. Compared with many other sectors, agriculture and forestry are less attractive for the younger generation. This means that the supply of human resources for farming may continue to decrease in the foreseeable future. The development of intelligent vehicles can serve to preserve some farming expertise. As a consequence to labour shortage, labour costs are rising, especially to highly skilled people.
Another aspect is avoidance of operator health problems from dusts and chemicals, which today is limited by various precautions as inconvenient breading filter and skin protecting devices.
1.2 Environmental impact
The most important environmental factors affected by a shift to small autonomous machines are soil compaction, energy savings, emission and leaching of pesticides and nutrients.
1.2.1 Soil compaction
Even though, many tractors are fitted with broad tyres , which producing less compaction in the topsoil, there is always a consequential problem for a severe deep compaction under the topsoil. According to numerous investigations this is beginning to take place when axle loads exceeds 6 ton. Fear is accelerated if we bear in mind that the latest tractors with attached implements can weigh up to 18 tons.
Schønning et al. (2000) states, that deep soil compaction will in practice be permanent and cannot be repaired by natural processes neither by known growing processes. Permanent yield loss has been measured to 3-5%. In reality yield loss up to 15% may be expected.
According to Håkansson (2000), soil compaction increases the need of tillage as well as the draft of tillage implements. Therefore soil compaction increases the costs, the energy requirement as well as the environmental effects of tillage. Various investigations he refer to has shown, that this increase may be 300 % or more, when heavy machinery is operated with single wheel configuration in wet soil condition, while it may be insignificant when the same machinery is operated with double wheel configuration in dry soil conditions.
Already Nielsen et al., (1977) in their motivation for researching possibilities for development of a master and slave tractor system mentions versatility and avoiding deep soil compaction as reasons for using smaller tractors
1.2.2 Energy savings
Another factor of significant importance is the reduction of energy inputs. It has been stated by many researchers , that up to 80-90% of the energy that goes into traditional tillage, is required for repairing the compaction caused by the large tractors. Another factor is, that the heavy equipment uses a quite large amount to.
1.2.3 Site and plant specific treatment
In agriculture new technology is being developed for site and plant specific treatments. This technology uses many technological improvements in the area of electronics, computers and sensors. Global position system (GPS) provides precise satellite-based information on the location of various soil types, plants and machinery. Remote sensing devices offer the potential of precise mapping of soil and crop conditions. Direct sensing technology can monitor ambient field conditions at fixed location. Electronic control systems facilitate precise placement of measured quantities of seed, fertiliser, herbicides and water (Blackmore, 1994). Additionally, developments in communications and computers provide the basis for rapid transfer of data.
Together with this spatially management strategies are being developed, which attempt to approach the varied agronomic operations that take place, starting with primary tillage and ending with crop harvesting, (Earl, 2000).
The mentioned processes are diverse in type including maps of field operations and final yields, remotely sensed images and records of actual treatment. When agronomic control functions are combined with navigation systems for automatic guidance, large data exchange and processing requirements result such that conventional analogue systems must be exchanged for electronic systems. The majority of the data and information required for autonomous field operations are position and time related. The use of the Geographic Information Systems (GIS) that appears pivotal to the development of an integrated approach that may ultimately provide the basis for implementing autonomous field operations.
Being able to incorporate all of these new technologies in agriculture, leads to a new scale approach. Presently the mostly used strategy is the field scale approach, where each field gets its specific treatment. The next strategy is the sub-field approach, where each sub-field of certain homogeneity gets its specific treatment. The future approach may be the plant scale strategy, which is called "phytotechnology" by the Japanese.
The site or plant specific treatments being possible by this technology means that blanket approach of herbicide and fertilizer applications on the whole field may be replaced by applications adapted to the local requirements and reduce the amounts used considerable. By doing that, we will move a long step towards sustainable and ecological agriculture.
1.2.4 Added value
Apart from the above advantages the data collected by the sensor systems of autonomous or precision machinery or may be used to improve management in other ways, i.e. estimation of crop quality and yield values for sales decisions as well as identification of problem areas or plants of the field.