Each of these steps takes time, and economic models say little about how much time. Other activities, like changes in study habits, could be quicker but, again, economic models say nothing about the length of time. Will children change their study habits as soon as they get electricity, or will they need an adjustment period before they reach a new equilibrium? Habit formation is not immediate, but in addition study time is jointly determined with time allocated to other activities. Time allocated to those other activities may be fluctuating in response to changes in their parents’ time allocation and, as we have just discussed, these changes take time. The time-resilience of the effects is another important aspect of the electrification effects, and regrettably no solid conclusions can be derived from our study. The problem is that the size and significance of an effect may change between rounds because the effect is temporary, or because the electrification rate in the non-encouraged group is catching up with that in the encouraged group, so the non-encouraged group is experiencing the same effects as the encouraged group. Think of a regression of labor supply on voucher allocation, estimated round by round. Assume we find a positive and significant coefficient in the first follow-up survey, but non-significant coefficients in the regressions for the second and third follow-ups. The increase in labor supply may have been a transitory effect of electrification. For instance, think of households increasing their labor supply to raise the money for the connection fee and some appliances, after which labor supply reverts to the preelectrification level. However, it may also be the case that the effect is permanent, but since with the pass of time the electrification rate in the non-encourage group catches up with the encouraged group, non-encouraged households are starting to experience the same changes than their counterparts4 . As the electrification rate increases in the non-encouraged group, these households will experience similar changes as the treatment group: labor supply will increase among the non-encouraged group, catching up with the encouraged group, and the coefficient on voucher allocation would become non-significant. To distinguish empirically between “fading out” and “catching up” effects,horticulture products one could think of looking at the evolution of the mean outcome per group. If the difference in labor supply disappears because the mean in the treatment group reverted to the pre-electrification level, one may be tempted to label the effect as “fading out”.
If the difference disappears because the mean in the non-encouraged group jumped closer to the encouraged group, one may be tempted to label the effect as “catching up”, but this issue is not as simple. Imagine that electrification increases labor supply 5 percentage points in the first year and then labor supply reverts to its baseline level. By the first follow-up survey we would find this difference between encouraged and non-encouraged households. Imagine that between the first and second follow up there is an economy wide shock that increases labor supply by 5 percentage points. The effect of electrification fades out in the encouraged group, decreasing labor supply by 5 percentage points, but the shock counteracts this effect, so the encouraged group ends up with the same level as in the first follow-up, 5 percentage points above their baseline value. Due to the economy-wide shock, labor supply also increases by 5 percentage points among the non-encouraged group. Since labor supply is higher for both groups, it would appear that the non-encouraged group “caught up” with the encouraged group, while in reality the effect in labor supply just faded out. It is similarly easy to come up with examples showing effects that seem to fade out when in reality groups are catching up. Therefore, to differentiate catching up from fading out we would need an experimental group that remains off the grid during the whole study period.We must significantly increase the production of consumer-safe, high-quality food, feed, fiber and bio-fuel products to cover the needs of an increasing world population that has more purchasing power and affluence, and ensuing per capita consumption. This must be accomplished in an economically and environmentally sustainable fashion that conserves the resource base, including biodiversity, water, and soil, despite limitations in arable land and fresh water resources. On the cultivation side, agricultural robotics technologies are essential in achieving this goal by providing mobile sensing, computation and actuation that enable precision farming at ever – increasing spatial and temporal resolutions, even at individual plant level. Such selective, individual plant care systems have been called “phytotechnology” and hold great potential for maximizing production while minimizing water, chemical and energy inputs. On the breeding side, fast development of radically improved crop varieties will rely on our ability to functionally link – to model and predict – the plant phenotype as the result of the interactions of genotype, field environment, and crop management.
This is the challenging task of field phenomics or phenotyping, i.e., the automated, high–throughput, proximal, non–destructive measurement of plants’ phenotypes in fields. “Breeding is essentially a numbers game: the more crosses and environments used for selection, the greater the probability of identifying superior variation” . Agricultural robots can offer the mobility, advanced sensing and physical sampling required for high-throughput field phenotyping.In the past decades, farmers, and in particular fruit, vegetable and horticultural farmers have relied on hired, low-wage workers, especially during the harvest periods. Recent studies indicate that as a result of socioeconomic, structural and political factors, local and migrant farm labor supply cannot keep up with demand in many parts of the world . Also, due to increasing industrialization and urbanization large countries like China are already moving towards the Lewis turning point, where surplus rural labor reaches a financial zero ; China is expected to reach it between 2020 and 2025 . Agricultural robots hold the potential to remedy existing and imminent farm labor shortages by increasing worker efficiency and safety acting as co-bots interacting with workers , or by replacing workers in low skill, labor-intensive tasks, like manual weeding or fruit and vegetable harvesting.Many agricultural robots have been developed to perform precision farming operations and replace or augment humans in certain tasks. These robots come in two main types: I) self-propelled mobile robots, and II) robotic “smart” implements that are carried by a vehicle. Type-I robots span wide ranges of sizes and designs. Conventional agricultural self propelled machines such as tractors, sprayers, and combine harvesters have been “robotized” over the last decade through the introduction of GPS/GNSS auto-guidance systems. These machines are commercially available today and constitute the large majority of “agricultural robots”. They can drive autonomously in parallel rows inside fields while a human operator supervises and performs cultivation-related tasks; turn autonomously at field headlands to enter the next row; and coordinate their operations . Autonomous cabinless general purpose ‘tractor robots’ were recently introduced by several companies that are compatible with standard cultivation implements . These larger robots are designed primarily for arable farming related operations that require higher power and throughput, such as ploughing, multi-row seeding, fertilizing, and spraying, harvesting and transporting.
A large number of smaller type-I special purpose mobile robots have also been introduced for lower-power applications such as scouting and weeding of a smaller number of rows at a time. Most of these robots are research prototypes introduced by various research groups. A few commercial or near-commercial mobile robots have emerged in applications like container handling in nurseries and seeding , respectively. Small robots like Xaver are envisioned to operate in teams and are an example of a proposed paradigm shift in the agricultural machinery industry,plant grow trays which is to utilize teams of small lightweight robots to replace large and heavy machines, primarily to reduce soil compaction.Type-II robots have been developed for various applications, and some are already commercially available, in applications like transplanting, lettuce thinning and mechanical weeding . Robotic implements at pre-commercial stage are also developed for applications like fruit harvesting and vine pruning in orchards and vineyards, respectively. Other orchard operations such as flower and green fruit thinning to control crop load have also been targeted for automation.Recent review articles have discussed some of the opportunities and challenges for agricultural robots and analyzed their functional sub-systems ; summarized reported research grouped by application type and suggested performance measures for evaluation ; and presented a large number of examples of applications of robotics in the agricultural and forestry domains and highlighted existing challenges . The goals of this article are to: 1) highlight the distinctive issues, requirements and challenges that operating in agricultural production environments imposes on the navigation, sensing and actuation functions of agricultural robots; 2) present existing approaches for implementing these functions on agricultural robots and their relationships with methods from other areas such as field or service robotics; 3) identify limitations of these approaches and discuss possible future directions for overcoming them. The rest of the article is organized as follows. The next section discusses autonomous navigation , as it is the cornerstone capability for many agricultural robotics tasks. Afterwards, sensing relating to crop and growing environment is discussed, where the focus is on assessing information about the crop and its environment in order to act upon it. Finally, interaction with the crop and its environment is discussed, followed by summary and conclusions. Clearly, the first three operations are not independent. For example, the spatial arrangement of field rows and row-traversal sequence that minimize working time depend not only on field geometry and row spacing, but also on vehicle mobility and maneuvering during turning at headlands to switch rows. The prevailing approach has been to assume obstacle-free headlands and use geometric approximations of headland maneuvering costs derived analytically – rather than numerically – to solve problems #1 or #2 independently, or combined. In the general robotics literature the combined problem is referred to as coverage path planning . An emerging idea in agricultural robotics is the utilization of teams of small autonomous machines to replace large machines . In such scenarios, routing, motion planning and auto-guidance approaches must be extended to multiple robots.
When these machines operate in parallel but independently the extensions deal mostly with splitting the field and avoiding collisions. However, when machines collaborate, as for example combine harvesters and unloading service trucks do during harvesting, issues of coordination, scheduling and dispatching need to be addressed. This scenario is also known as field logistics and will be covered as part of vehicle routing.The operation computes a complete spatial coverage of the field with geometric primitives that are compatible with and sufficient for the task, and optimal in some sense. Headland space for maneuvering must also be generated. Agricultural fields can have complex, non-convex shapes, with non-cultivated pieces of land inside them. Fields of complex geometry should not be traversed with a single orientation; the efficiency would be too low because of excessive turning. Also, fields are not necessarily polygonal, they may have curved boundaries and may not be flat. Additionally, most agricultural machines are nonholonomic and may carry a trailer/implement, which makes computing turning cost between swaths non trivial . Finally, agricultural fields are not always flat and field traversal must take into account slope and vehicle stability and constraints such as soil erosion and compaction.Computing a complete spatial coverage of a field with geometric primitives is in principle equivalent to solving an exact cellular decomposition problem .Choset and Pignon, developed the Boustrophedon cellular decomposition . This approach splits the area into polygonal cells that can be covered exactly by linear back-and-forth motions. Since crops are planted in rows, this approach has been adopted by most researchers. A common approach is to split complex fields into simpler convex sub-fields via a line sweeping method, and compute the optimal driving direction and headland arrangement for each sub-field using an appropriate cost function that encodes vehicle maneuvering in obstacle-free headland space . This approach has been extended for 3D terrain .Existing approaches assume that headland space is free of obstacles and block rows are traversed consecutively, i.e., there is no row-skipping.