Reliable Trajectories for Dynamic Quadrupeds using Analytical Costs and Learned Initializations

Melon, Oliwier; Geisert, Mathieu; Surovik, David; Havoutis, Ioannis; Fallon, Maurice

doi:10.1109/icra40945.2020.9196562

Cited by 27 publications

(29 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The effect of the cost on the wheel magnitude difference is illustrated in Figure 2. Similar to our findings, [23] also reports introducing costs in the optimization improves the quality of the planned motion. Unfortunately, improved quality comes with more computation time (about 2x decrease of the RT factor).…”

Section: B Extensionssupporting

confidence: 91%

Terrain-Adaptive Planning and Control of Complex Motions for Walking Excavators

Jelavić

Berdou

Jud

et al. 2020

2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

View full text Add to dashboard Cite

This article presents a planning and control pipeline for legged-wheeled (hybrid) machines. It consists of a Trajectory Optimization based planner that computes references for end-effectors and joints. The references are tracked using a whole-body controller based on a hierarchical optimization approach. Our controller is capable of performing terrain adaptive whole-body control. Furthermore, it computes both torque and position/velocity references, depending on the actuator capabilities. We perform experiments on a Menzi Muck M545, a full size 31 Degrees of Freedom (DoF) walking excavator with five limbs: four wheeled legs and an arm. We show motions that require full-body coordination executed in realistic conditions. To the best of our knowledge, this is the first work that shows the execution of whole-body motions on a full size walking excavator, using all DoFs for locomotion.

show abstract

Section: B Extensionssupporting

confidence: 91%

Terrain-Adaptive Planning and Control of Complex Motions for Walking Excavators

Jelavić

Berdou

Jud

et al. 2020

2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

View full text Add to dashboard Cite

show abstract

“…In this work, we utilize the trajectory optimizaztion solver TOWR [4] together with the smoothing costs described in [6] to generate base and end-effector (feet) motion plans. The TOWR framework formulates the locomotion problem of legged robots as a non-linear optimization problem which can generate dynamic trajectories for locomotion over complex 3D terrains.…”

Section: A Trajectory Optimizationmentioning

confidence: 99%

“…Despite its minimal parameterization, the computational performance of such methods are not yet suitable for fast online motion planning. In this regard, much of the research focuses on either improving the convergence rate of this formulation by exploiting learnt initial guess [5], [6] or by introducing feasibility constraints in order to account for the dynamic limits of the actual hardware [7]. In both these cases, the successful execution of the pre-determined motion plan relies Fig.…”

Section: Introductionmentioning

confidence: 99%

Real-Time Trajectory Adaptation for Quadrupedal Locomotion using Deep Reinforcement Learning

Gangapurwala

Geisert

Orsolino

et al. 2021

2021 IEEE International Conference on Robotics and Automation (ICRA)

Self Cite

View full text Add to dashboard Cite

We present a control architecture for real-time adaptation and tracking of trajectories generated using a terrain-aware trajectory optimization solver. This approach enables us to circumvent the computationally exhaustive task of online trajectory optimization, and further introduces a control solution robust to systems modeled with approximated dynamics. We train a policy using deep reinforcement learning (RL) to introduce additive deviations to a reference trajectory in order to generate a feedback-based trajectory tracking system for a quadrupedal robot. We train this policy across a multitude of simulated terrains and ensure its generality by introducing training methods that avoid overfitting and convergence towards local optima. Additionally, in order to capture terrain information, we include a latent representation of the height maps in the observation space of the RL environment as a form of exteroceptive feedback. We test the performance of our trained policy by tracking the corrected set points using a model-based whole-body controller and compare it with the tracking behavior obtained without the corrective feedback in several simulation environments. We also show successful transfer of our training approach to the real physical system and further present cogent arguments in support of our framework.

show abstract

“…However, these approaches usually suffer from high computational time hence they are often restricted to offline (open-loop) use. In general, offline planners [10,11] neither adapt to quick terrain changes nor cope with state drifts and uncertainties. To address this issue the concept of online re-planning can be used.…”

Section: Introductionmentioning

confidence: 99%