Introduction
-
on rough terrain depending on the mission requirements.
-
on heterogeneous terrains.
-
in non-smooth environment.
Related works
Sampling based algorithms
Geometric analytic approaches
Graph search algorithms
Model based algorithms
Uncertainty-aware planning
Qualitative comparison
Methods
Assumptions
Hopper Conditions
Markov decision processes
Payoff function
Simulation study
Modelling assumptions
Parameters | Values | |||||||
---|---|---|---|---|---|---|---|---|
i = 1 | i = 2 | i = 3 | i = 4 | i = 5 | i = 6 | i = 7 | i = 8 | |
\(E_i[x]\) | 1 | 5 | 5 | 7 | 8 | 10 | 12 | 15 |
\(E_i[y]\) | 6 | 0 | 20 | 14 | 8 | 18 | 1 | 10 |
\(\sigma _{i, x}\) | 2 | |||||||
\(\sigma _{i, y}\) | 2 | |||||||
\(\sigma _{i, xy}\) | 0 |
Action name | Directions | |||
---|---|---|---|---|
Moving orientations | North | South | East | West |
Turning orientations | Turn right | Turn left | ||
Vertical hop orientations | North | South | East | West |
Results and discussion
Simulations on hard ground
-
can reach to the expected point with a \(70\left( 1-\frac{\theta _\mathrm {front}}{(\pi /6)} \right)\) % possibility
-
may not change the state with a \(70\frac{\theta _\mathrm {front}}{(\pi /6)}\)% possibility
-
may be off to the left with a \(15\left( 1-\frac{\theta _\mathrm {right}}{(\pi /6)} \right)\)% possibility or to the right with a \(15\left( 1-\frac{\theta _\mathrm {left}}{(\pi /6)} \right)\)% possibility
Simulations on heterogeneous terrain
-
can reach to the expected point with a \(50\left( 1-\frac{\theta _\mathrm {front}}{(\pi /6)} \right)\)% possibility
-
may not change the state with a 50\(\frac{\theta _\mathrm {front}}{(\pi /6)}\) possibility
-
might get a stuck with a 20% possibility
-
may be off to the left with a \(15\left( 1-\frac{\theta _\mathrm {right}}{(\pi /6)} \right)\)% possibility or to the right with a \(15\left( 1-\frac{\theta _\mathrm {left}}{(\pi /6)} \right)\)% possibility
Simulations on sandy terrain
Vertical hopping simulation
-
can reach to the expected point with a \(70 \left( 1-\frac{\Delta h}{h_\mathrm {max}} \right) \%\) possibility
-
may be off to the left with a \(15 \left( 1- \frac{\Delta h}{h_\mathrm {max}} \right) \%\) possibility or to the right with a \(15\left( 1- \frac{\Delta h}{h_\mathrm {max}} \right) \%\) possibility
-
fails to get on the rock with a \(100 \frac{\Delta h}{h_\mathrm {max}} \%\) possibility