This paper focuses on the problem of detecting and reacting to changes in the distribution of a sensorimotor controller’s observables. The key idea is the design of policies that can take conformal quantiles as input, to detect distribution shifts with formal statistical guarantees, which we define as conformal policy learning. We show how to design such policies by using conformal quantiles to switch between base policies with different characteristics, e.g. safety or speed, or directly augmenting a policy observation with a quantile and training it with reinforcement learning. Theoretically, we show that such policies achieve the formal convergence guarantees in finite time. In addition, we thoroughly evaluate their advantages and limitations on two use cases: simulated autonomous driving and active perception with a physical quadruped. Empirical results demonstrate that our approach outperforms five baselines. It is also the simplest of the baseline strategies besides one ablation. Being easy to use, flexible, and with formal guarantees, our work demonstrates how conformal prediction can be an effective tool for sensorimotor learning under uncertainty.

Polyculture farming has environmental advantages but requires substantially more labor than monoculture farming. We present novel hardware and algorithms for automated pruning and irrigation. Using an overhead camera to collect data from physical$1.5~m^{2}$garden testbeds, the autonomous system utilizes a learned Plant Phenotyping convolutional neural network and a Bounding Disk Tracking algorithm to evaluate the individual plant distribution and estimate the state of the garden each day. From this garden state, AlphaGardenSim selects plants to autonomously prune. A trained neural network detects and targets specific prune points on the plant. Two custom-designed pruning tools, compatible with a FarmBot commercial gantry system, are experimentally evaluated. Irrigation is automated using soil moisture sensors. We present results for four 60-day garden cycles. Results suggest the system can autonomously achieve 94% normalized plant diversity with pruning shears while maintaining an average canopy coverage of 84% by the end of the cycles. For code, videos, and datasets, see https://sites.google.com/berkeley.edu/pruningpolyculturej/home.Note to Practitioners—While polyculture farming is closer to how plants grow in nature, it is considered more labor intensive that monoculture farming. In this paper we present approaches and custom hardware for automation of pruning and irrigation. Physical experiments suggest that automation can yield both high coverage and diversity.

The AlphaGarden is an automated testbed for indoor polyculture farming which combines a first-order plant simulator, a gantry robot, a seed planting algorithm, plant phenotyping and tracking algorithms, irrigation sensors and algorithms, and custom pruning tools and algorithms. In this paper, we systematically compare the performance of the AlphaGarden to professional horticulturalists on the staff of the UC Berkeley Oxford Tract Greenhouse. The humans and the machine tend side-by-side polyculture gardens with the same seed arrangement. We compare performance in terms of canopy coverage, plant diversity, and water consumption. Results from two 60-day cycles suggest that the automated AlphaGarden performs comparably to professional horticulturalists in terms of coverage and diversity, and reduces water consumption by as much as 44%. Code, videos, and datasets are available at https//sites.google.com/berkeley.edulsystematiccomparison

We consider a decluttering problem where multiple rigid convex polygonal objects rest in randomly placed positions and orientations on a planar surface and must be efficiently transported to a packing box using both single and multi-object grasps. Prior work considered frictionless multi-object grasping. In this paper, we introduce friction to increase the number of potential grasps for a given group of objects, and thus increase picks per hour. We train a neural network using real examples to plan robust multi-object grasps. In physical experiments, we find a 13.7% increase in success rate, a 1.6x increase in picks per hour, and a 6.3x decrease in grasp planning time compared to prior work on multi-object grasping. Compared to single-object grasping, we find a 3.1x increase in picks per hour.

Polyculture farming, where multiple crop species are grown simultaneously, has potential to reduce pesticide and water usage while improving the utilization of soil nutrients. However, it is much harder to automate polyculture than monoculture. To facilitate research, we present AlphaGardenSim, a fast, first order, open-access polyculture farming simulator with single plant growth and irrigation models tuned using real world measurements. AlphaGardenSim can be used for policy learning as it simulates inter-plant dynamics, including light and water competition between plants in close proximity and approximates growth in a real greenhouse garden at 25,$000\times $the speed of natural growth. This paper extends earlier work with a new action space that includes planting, which dynamically finds new seed locations that increases resources utilization, and an adaptive sampling technique to reduce the number of actions taken at each timestep without affecting performance. We also evaluate other automation policies using a novel metric that combines plant diversity and canopy coverage. Code and supplementary material can be found athttps://github.com/BerkeleyAutomation/AlphaGarden.Note to Practitioners—Monoculture farming is often characterized by heavy agrichemical inputs, such as chemical fertilizers and pesticides, and increased vulnerability to disease and pestilence. This paper is motivated by the lack of long-term sustainability of industrial agriculture, and its implications for human food security. Although polyculture is a sustainable alternative to monoculture farming, it requires more human labor and is more challenging to automate. In this paper we propose a fast, first order simulator that simulates the growth of plants in a polyculture setting. Simulation experiments suggest that the simulator can be used to learn a planting, watering and pruning plan a robot can follow to produce maximal yield from a diverse set of plants with limited irrigation, however it has not yet been tested on a physical garden. In future research we will develop a fully automated controller that will operate planting, irrigation and pruning tools in a physical garden over multiple plant growth cycles.

The difficulty in specifying rewards for many real-world problems has led to an increased focus on learning rewards from human feedback, such as demonstrations. However, there are often many different reward functions that explain the human feedback, leaving agents with uncertainty over what the true reward function is. While most policy optimization approaches handle this uncertainty by optimizing for expected performance, many applications demand risk-averse behavior. We derive a novel policy gradient-style robust optimization approach, PG-BROIL, that optimizes a soft-robust objective that balances expected performance and risk. To the best of our knowledge, PG-BROIL is the first policy optimization algorithm robust to a distribution of reward hypotheses which can scale to continuous MDPs. Results suggest that PG-BROIL can produce a family of behaviors ranging from risk-neutral to risk-averse and outperforms state-of-the-art imitation learning algorithms when learning from ambiguous demonstrations by hedging against uncertainty, rather than seeking to uniquely identify the demonstrator’s reward function.

Polyculture farming is a sustainable farming technique based on synergistic interactions between differing plant types that make them more resistant to diseases and pests and better able to retain water. Reduced uniformity can reduce use of pesticides, fertilizer, and water, but is more labor intensive and more challenging to automate. We describe a scaled physical testbed (1.5m×3.0m) that uses a high resolution camera and soil sensors to monitor polyculture plants to facilitate tuning of plant growth, companion effects, and irrigation parameters for a first-order garden simulator. We use this simulator to develop a novel seed placement algorithm that increases coverage and diversity, and a learned pruning policy. In simulation experiments, the seed placement algorithm yields 60% more coverage and 10% more diversity than random seed placement and the learned pruning policy runs 1000X faster than a procedural lookahead policy to achieve high leaf coverage and plant diversity on adversarial gardens that include plant species with diverse growth rates. These models and policies provide the groundwork for a fully-automated system under development. Code, datasets and supplementary material can be found at https://github.com/BerkeleyAutomation/AlphaGarden/.