# PLATO: Policy Learning using Adaptive Trajectory Optimization

arXiv: https://arxiv.org/abs/1603.00622 · HTML: https://mlsanity.com/paper/1603.00622/

## Key facts

- Cited by 24 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 2 Mar 2016.

## Built on by (most cited first, 24 of 24)

- [Following High-level Navigation Instructions on a Simulated Quadcopter with Imitation Learning](https://mlsanity.com/paper/1806.00047/) (2018)
- [CGD: Constraint-Guided Diffusion Policies for UAV Trajectory Planning](https://mlsanity.com/paper/2405.01758/) (2024)
- [Imitation Learning from MPC for Quadrupedal Multi-Gait Control](https://mlsanity.com/paper/2103.14331/) (2021)
- [Lipschitzness Is All You Need To Tame Off-policy Generative Adversarial Imitation Learning](https://mlsanity.com/paper/2006.16785/) (2020)
- [Demonstration-Efficient Guided Policy Search via Imitation of Robust Tube MPC](https://mlsanity.com/paper/2109.09910/) (2021)
- [On-Policy Robot Imitation Learning from a Converging Supervisor](https://mlsanity.com/paper/1907.03423/) (2019)
- [Towards Disturbance-Free Visual Mobile Manipulation](https://mlsanity.com/paper/2112.12612/) (2021)
- [Bootstrapping Motor Skill Learning with Motion Planning](https://mlsanity.com/paper/2101.04736/) (2021)
- [Constrained Imitation Learning for a Flapping Wing Unmanned Aerial Vehicle](https://mlsanity.com/paper/2206.03987/) (2022)
- [Output Feedback Tube MPC-Guided Data Augmentation for Robust, Efficient Sensorimotor Policy Learning](https://mlsanity.com/paper/2210.10127/) (2022)
- [Agile Reinforcement Learning through Separable Neural Architecture and Applications](https://mlsanity.com/paper/2601.23225/) (2026)
- [Learning Solution Manifolds for Control Problems via Energy Minimization](https://mlsanity.com/paper/2203.03432/) (2022)
- [Distillation Policy Optimization](https://mlsanity.com/paper/2302.00533/) (2023)
- [Tube-NeRF: Efficient Imitation Learning of Visuomotor Policies from MPC using Tube-Guided Data Augmentation and NeRFs](https://mlsanity.com/paper/2311.14153/) (2023)
- [Angle Robustness Unmanned Aerial Vehicle Navigation in GNSS-Denied Scenarios](https://mlsanity.com/paper/2402.02405/) (2024)
- [SUN: Agentic Robot Policy Learning with Persistent Task Programs](https://mlsanity.com/paper/2608.31167/) (2026)
- [Formal Policy Learning from Demonstrations for Reachability Properties](https://mlsanity.com/paper/1903.00589/) (2019)
- [MPC-guided Imitation Learning of Neural Network Policies for the Artificial Pancreas](https://mlsanity.com/paper/2003.01283/) (2020)
- [Leveraging Experience in Lazy Search](https://mlsanity.com/paper/2110.04669/) (2021)
- [Reinforcement Learning from Demonstrations by Novel Interactive Expert and Application to Automatic Berthing Control Systems for Unmanned Surface Vessel](https://mlsanity.com/paper/2202.11325/) (2022)
- [Neural-Progressive Hedging: Enforcing Constraints in Reinforcement Learning with Stochastic Programming](https://mlsanity.com/paper/2202.13436/) (2022)
- [TubeDAgger: Reducing the Number of Expert Interventions with Stochastic Reach-Tubes](https://mlsanity.com/paper/2510.00906/) (2025)
- [GCImOpt: Learning efficient goal-conditioned policies by imitating optimal trajectories](https://mlsanity.com/paper/2604.22724/) (2026)
- [Amortising Trajectory Optimisation for Residual MPC via Implicit Contact Differentiation](https://mlsanity.com/paper/2607.24959/) (2026)

## Abstract (arXiv)

Policy search can in principle acquire complex strategies for control of robots and other autonomous systems. When the policy is trained to process raw sensory inputs, such as images and depth maps, it can also acquire a strategy that combines perception and control. However, effectively processing such complex inputs requires an expressive policy class, such as a large neural network. These high-dimensional policies are difficult to train, especially when learning to control safety-critical systems. We propose PLATO, an algorithm that trains complex control policies with supervised learning, using model-predictive control (MPC) to generate the supervision, hence never in need of running a partially trained and potentially unsafe policy. PLATO uses an adaptive training method to modify the behavior of MPC to gradually match the learned policy in order to generate training samples at states that are likely to be visited by the learned policy. PLATO also maintains the MPC cost as an objective to avoid highly undesirable actions that would result from strictly following the learned policy before it has been fully trained. We prove that this type of adaptive MPC expert produces supervision that leads to good long-horizon performance of the resulting policy. We also empirically demonstrate that MPC can still avoid dangerous on-policy actions in unexpected situations during training. Our empirical results on a set of challenging simulated aerial vehicle tasks demonstrate that, compared to prior methods, PLATO learns faster, experiences substantially fewer catastrophic failures (crashes) during training, and often converges to a better policy.